IMO it isn't so important to have the FMs perfect as it is to have the relative strengths and weaknesses modelled correctly. I.e. If a spitfire can outturn a p51 at 200mph by a large margin, then the planes are historically correct. For all the reasons above it is impossible to make the FM 100% accurate. I read the turn rate thread for a while and got thinking about all of the factors involved in calculating how fast a plane should be able to turn 360 degrees. Were all the planes tested at corner speed? What about altitude? Temperature? Octane of fuel? New airplane or well worn capured test example? How much fuel, ammo, flap settings? I'll bet if you add up all these factors you will have a large possibility for error from one plane to another. I'm also curious to see what effect prop drag will have on turn rates. It seems to me that cutting the throttle could help you corner faster for 90 degrees or so if you were above corner speed.
To me AH "feels right." I used to fly MSCFS a little bit and it seemed really arcade like. You could turn a p51 in circles for ever and you wouldn't stall. Now I've never flown a real plane, but to me a p51 being able to continuously pull 4 Gs as long as you want seems impossible.
The only complaint that I have with any of the plane FMs right now is that the p51 can lift its tailwheel off the ground at 5 mph when rolling by hitting the brakes. (It then usually breaks off when it hits the ground.

)
------------------
bloom25
THUNDERBIRDS