That looks okay to me, but it’s possible that MSFS has a different ozone distribution than the one I calculated the coefficients for, which would mean not enough ozone and hence too much green. You can try multiplying them with 1.5 or something to compensate for it.
Another issue is that multiple scattering doesn’t seem to be enough (which is an issue in XP12 too, they boost multiple scattering artificially because of that), which could improve the results as well. But as I don’t know what the original value of the multiple scattering multiplier is (or even if it exists), there is no way for me to change it.