Hi Justin,
firstly, how do you determine the difference in quality of the mesh? 
In my eyes, they are almost identical. The differences are certainly less than a good and a perfect alignment would show, so it is not neccessarily due to the adjustments of the images. Or did you use the same alignment and only swapped the images?
Anyway, as far as I have gleaned from my own research on the topic, changing the images is fine as long as you do not introduce severe artifacts. That means brigthening shadows and reducing highlights should be ok as long as you don’t lose any brightness information. However, I would not chop the histogram very much since that can (afaik) change the pattern or feature recognition of photogrammetry. Then again, every project is different and what might be good for you might be bad for somebody (or something) else.
I do not know of any RC specific guideline, but I think that the general process is not different from other photogrammetry software, so other guidelines should be transferable.