The powerplant is well understood, suggesting a mature/pre-existing concept (eg a "conventional" nuclear fission reactor):

The powerplant is well understood, suggesting a mature/pre-existing concept (eg a "conventional" nuclear fission reactor) rather than an exotic new concept:

New power/propulsion concepts would probably fall outside the scope of this study, which sounds fairly small and limited to what the sponsor could potentially accomplish in the short/medium term.

Note: "Conventional" is used in the same sense as the OP (I hope): "conventional reactor based on fissionable material." I do not use it here to refer to a specific reactor generation, fuel type, fuel phase, or coolant type.

Citation 65 in the question does reference a nuclear-powered Global Hawk:

But this August 2003 page appears to be copied from an earlier February 2003 article in New Scientist:

This is the only reference to a nuclear-powered Global Hawk I've found. I'm not sure this is a reliable source, however, for two reasons. First, this article was published in 2003, just five years after the Global Hawk started flying. That's not a lot of time for AFRL to conduct two feasibility studies for nuclear-powered Global Hawks, though not impossible. I've been unable to find these studies. It's possible these studies were for nuclear-powered ISR drones in general, not specifically for the Global Hawk. Second, the New Scientist article references a problematic study by Carl Collins on using a metastable nuclear isomer of Hafnium (Hafnium-178m2) as a gamma/heat source by using xrays to induce the transition/gamma release. The claim has since been discredited, (though this may not have been apparent at the time of publication). I'm not saying the author is unreliable (he seems like a decent STEM writer for the MIT Tech Review and Scientific American), but he doesn't seem to have a background in defense or aviation. So there's a possibility he conflated "Global Hawk" with "drone," which wouldn't be unusual of even good journalists then or now.

Northrop Grumman (maker of the Global Hawk) did sponsor Sandia to study the feasibility of using 'non-hydrocarbon' fuels for "ultrapersistence" (read: nuclear). 2012 Report summary. Quotes:

Upon completion of the CRADA, Sandia provided a final out-brief to NGIS UMS [Northrop Grumman Integrated Systems Unmanned Systems]. Accomplishments: The effort concentrated on propulsion and power technologies that went well beyond existing hydrocarbon technologies. It contrasted and compared eight heat sources technologies, three power conversion, two dual cycle propulsion system configurations [indirect cycle?], and a single electrical power generation scheme [to power the ever-hungrier avionics mentioned earlier]. Overall performance, specific power parameters, technical complexities, security, safety, and other operational features were successfully investigated. Large and medium sized UAV systems [probably Global Hawk or X-47B and larger] were envisioned and operational flight profiles were developed for each concept. Heat source creation and support challenges for domestic and expeditionary operations were considered. Fundamental cost driver analysis was also performed. System development plans were drafted in order to determine where the technological and programmatic critical paths lay. NGIS UMS and SNL felt that the technical goals for the project were accomplished. NGIS UMS was quite pleased with the results of analysis and design although it was disappointing to all that the political realities would not allow use of the results.

Citation 65 (August 2003?) in the question does reference a nuclear-powered Global Hawk:

But this August 2003 page (above) appears to be copied from an earlier February 2003 article in New Scientist (below):

This is the only reference to a nuclear-powered Global Hawk I've found. I'm not sure this is a reliable source, however, for two reasons. First, this article was published in 2003, just five years after the Global Hawk started flying. That's not a lot of time for AFRL to conduct two feasibility studies for nuclear-powered Global Hawks, though not impossible. I've been unable to find these studies. It's possible these studies were for nuclear-powered ISR drones in general, not the Global Hawk specifically. Second, the New Scientist article references a problematic study by Carl Collins on using a metastable nuclear isomer of Hafnium (Hafnium-178m2) as a gamma/heat source by using xrays to induce the transition/gamma release. The claim has since been discredited (though this may not have been apparent at the time of publication). I'm not saying the author is unreliable (he seems like a decent STEM writer for the MIT Tech Review and Scientific American), but he doesn't seem to have a background in defense or aviation. So there's a possibility he conflated "Global Hawk" with "drone," which wouldn't be unusual of even good journalists then or now.

Northrop's nuclear-powered UAV work seems to predate Global Hawk:

Northrop Grumman is known to have patented a drone equipped with a helium-cooled nuclear reactor as long ago as 1986, and has previously worked on nuclear projects with the US air force research laboratory. Designs for nuclear-powered aircraft are known to go back as far as the 1950s. The Guardian

Northrop Grumman (maker of the Global Hawk) did sponsor Sandia to study the feasibility of using 'non-hydrocarbon' fuels for an "ultrapersistent" (read: nuclear) UAV. (Note that even this report doesn't mention the Global Hawk.) 2012 Report summary. Quotes:

Upon completion of the CRADA, Sandia provided a final out-brief to NGIS UMS [Northrop Grumman Integrated Systems Unmanned Systems]. Accomplishments: The effort concentrated on propulsion and power technologies that went well beyond existing hydrocarbon technologies. It contrasted and compared eight heat sources technologies, three power conversion, two dual cycle propulsion system configurations [indirect cycle?], and a single electrical power generation scheme [to power the ever-hungrier avionics mentioned earlier]. Overall performance, specific power parameters, technical complexities, security, safety, and other operational features were successfully investigated. Large and medium sized UAV systems [probably Global Hawk or X-47B and larger] were envisioned and operational flight profiles were developed for each concept. Heat source creation and support challenges for domestic and expeditionary operations were considered. Fundamental cost driver analysis was also performed. System development plans were drafted in order to determine where the technological and programmatic critical paths lay. NGIS UMS and SNL felt that the technical goals for the project were accomplished. NGIS UMS was quite pleased with the results of analysis and design although it was disappointing to all that the political realities would not allow use of the results.

Taking this assessment at face value, nuclear-powered UAVs sound technically feasible.

The powerplant is well understood, suggesting a mature/pre-existing concept (eg a "conventional" nuclear fission reactor):

Basic application of the advanced propulsion and power approach is well understood