Change Terminal Velocity formular to consider Buoyancy effect.

It seems like current SR2 doesn't model atmospheric buoyancy, considering this

The cube is about 4 times lighter than the Venusian atmosphere but it's still falling.

If calculating all the buoyancy effect is too burdening, how about overriding terminal velocity? If vertical velocity is lower than certain figure(saying, 3m/s) vertical velocity is overridden by terminal velocity with buoyancy formula. This sounds very bug-inducing tho.