I do notice that the parameters of stars aren't linked, for example, adjusting the mass of a star does not change the surface temperature. I guess that's just a WIP.
More noteworthy is the new galaxy simulation. Is the method of using two dimensional stellar cluster textures, instead of the much more realistic (yet less good looking in US1) method of using individual colored dust particles going to be continued? Will it become more dynamic and realistic in the coming versions? Seeing large groups of stars acting as one solid object seems to ruin the 'Simulation' in the 'galaxy simulation'.