This weekend I saw the Fly Baby at the Smithsonian Air & Space Museum at the Washington Dulles Airport, and it was at some point the smallest airplane (since surpassed by the Bumble Bee II, if I understand correctly). To be able to carry a pilot and lift the plane off the ground in the first place, but still be the smallest airplane, you are mostly limited by wingspan, or more generally, the total width of the airplane.
For the purpose of this question I would like to assume a given airfoil and to exclude modifications to the airfoil itself, i.e. any mechanism that increases lift in pure 2D flow, e.g. camber, slats, flaps, active blowing of the upper surface etc. I would like to limit the scope to the "3D planform" of the wing(s) - e.g. aspect ratio, sweep, depth, winglets/end plates/..., box wings, number of wings etc. The airplane needs to have a fuselage that holds the pilot and an engine. Otherwise the configuration can be whatever it needs to be. If the configuration for maximizing lift depends on the choice of airfoil, then just assume a suitable airfoil.
Both record-holding planes are biplanes, with the Bumble Bee having a relatively large horizontal stabilizer. It also has endplates on the wingtips, reducing lift loss due to the tip vortex.
How can one maximize the lift for a given width?
(I could list all of the ways that come to mind and ask whether and how much they help, but then I'd have "more than one question" and I'm not sure that would be suitable for this format)
Edit: Further clarification: I am asking about maximizing the coefficient of lift of the entire airplane, not about lift itself (which could be maximized by flying faster, to a certain extent). Also, I would like the answer to be valid for low flying speeds (incompressible flow).