Is it cost effective to put a motion system underneath a Full Flight Simulator? Yes, because it is a requirement for being able to conduct recurring training, type rating etc. A necessary cost for civil aviation, generating a potential revenue stream.
Why is it a requirement to use a motion system at all? Because of what @kevin says. Without direct stimulation of our inner ear acceleration sensors, the whole "flight" experience is just an interactive wide screen cinema set-up.
Do we need our inner ear to experience motion sensation? Yes and no. Give our brain a moment, and our steady state motion sensation is distilled quite accurately from our peripheral vision. Not from our central vision, where the instruments are.
But any changes in motion are immediately conveyed by our inner ear. That's why we have the sensors there: immediate, direct perception of acceleration. Not of velocity or position, which are integrated signals with the associated 90/180º phase lag, and undefined state of constants.
So our inventory of sensors for position and motion awareness is:
Central, stereoscopic vision for detailed perception of relative position. But stereoscopic vision is not applicable when reading flight instruments.
Peripheral vision for constant velocity detection (central vision is not very good at this, other than from stereoscopic distance perception).
Inner ear acceleration sensors for being alerted to instant changes of velocity.
Motion systems can do a pretty good job at reproducing direct accelerations, until the actuator stroke runs out of course. So for short term accelerations - sustained accelerations require a whole country as stroke. Gravity can be used to a limited extent for simulating sustained acceleration, with some pretty decent effects such as take-off push. An approach flown in a simulator with motion creates much more accurate results, even in the hands of an expert pilot, because the inner ear aids the interpretation from flight instrument data.
For controlling a helicopter in a hover, motion is absolutely essential. Trying to maintain an unstable equilibrium, like standing on top of a large inflatable ball, depends almost exclusively upon inner ear acceleration sensing.
What motion cannot do is reproduce increased load factors such as occurring in co-ordinated turns. Fighter simulators have g-seats and g-suits to do this. And for fighter simulators, it is not considered cost effective to use motion systems, because of the different focus in training tasks. The G-seats recreate the secondary effects from sustained load factor: the pilot is pressed into his seat, so the whole seat is lowered and the seat cushion pumped up. And the G-suit pumps air pressure to the pilots legs, what he experiences in the real aircraft to prevent the blood from being drained from the head.
Motion is pretty awesome, especially the modern electrical systems which are way faster than the old type hydraulic ones. Hydraulics are good for high static forces, but struggle to pump enough oil through the servo valves for fast actuator extensions. Brushless DC electric motors can exert high forces as well, produce instant accelerations, and at 4000 RPM have no problem with high actuator velocities.