With a large enough intercooler, it doesn't matter quite as much when the turbo is way out of its efficiency range. As long as the intake charge is cool enough to prevent detonation, you can still make more power (although when the knock limit drops below MBT (mean best timing/torque), you can sometimes do better by running more timing and less boost. Note that this is octane dependent - with race gas you can go much farther before you hit knock. The main problem without one is heat (less dense air, and knock). An intercooler is kind of like "cheating" by removing the "waste" heat generated by the turbo.
You can definitely say though, that a more efficient turbo at that boost pressure will produce *more* power because less power is wasted generating heat.
One way to think of it is that exhaust manifold backpressure is decreased and more air can flow through the engine (even though you're only changing the compressor, there is less flow resistance if the turbine is not as hard to turn). In a perfectly optimized turbo setup, the exhaust pressure can be lower than the intake boost pressure, and you can run stratospheric boost pressures (ie Formula 1 in the 80's). But at that boost level, with that turbo way out of the efficiency range, I wouldn't be surprised if the exhaust manifold pressure was 2-3 times the boost pressure, as the engine is trying to push exhaust in faster than it can escape through the turbine/wastegate.
The other effect of backpressure is that your manifold "boost" is less effective. When intake valve opens into the overlap phase (where the exhaust valve is still open), the pressure it's trying to push air into is that high manifold pressure. It's quite likely that some or all of the valve overlap at some higher RPMs, you're getting reversion back into the intake, until the piston starts going down, where it gets sucked back in. Even though you still have "boost", not as much air is actually making it into the cylinder before the intake closes. What you've done is replaced perfectly good air in the cylinder with inert, uncooled exhaust gas, just like an (uncooled) EGR valve would. Some auto manufacturers actually experiment with this kind of trick on purpose to meet emissions requirements, especially when the intake is in vacuum.
Also, even if the compressor is down below 55-60% efficiency, it's still more efficient than a roots blower. You can still produce more power, but just not very "efficiently"
The only real concern is that to produce that kind of boost and airflow (600 cfm), the turbo is spinning *way* beyond the shaft speed it was designed for. Make sure you're feeding full oil pressure to it (instead of the ~15psi from the head), but the turbo's life will still be shortened (yay shaft play)
The size of turbo that you "need" is very subjective. Many people run a setup that is not optimized, whether the turbo is "too small" or "too large". Since most tuners really don't care about fuel consumption at full throttle (unlike a racecars which have to go into the pit lane during a race to refuel), and a large number of them don't put in that much effort to really dial in things like lag, powerband, throttle response, and things that reduce laptimes, as long as they are tolerable, they can still end up reasonably "satisfied" (at least for a short time /ubbthreads/images/graemlins/rofl.gif)