I see where you're going with this but, in order to highlight the difference, perhaps it would have been more helpful to stretch each individually.
I hope you don't mind but I've just chucked your two images in PixInsight and done an auto stretch, with channels unlinked, to each. Of course this would be much better to do with the raws.
You can definitely see the much greater sensitivity to Ha that the modded camera has and even with just an auto stretch its colours are much closer to what we would consider 'correct'.
Nice one, I look forward to seeing how it performs with some other targets as you're getting some great detail out of your setup
