Well, the orange mask should be there to compensate for unwanted dye absorption, nothing else, but nothing's perfect. Orange mask + unwanted dye absorption together combined should become constant orange density. But on the other hand, the complete system is designed so that the first bit of the toe region is discarded in printing, i.e., it will become maximum black in print, so it may be possible that the mask responds differently in this area, not meant to appear in print anyway. Second, films are supposed to be processed in complete darkness so that real Dmin from film+process can be read somewhere in the film.
So maybe you are right; If this is the case, then the question is, do the prints (or properly adjusted scans for that matter) have this same cast in blacks? Or do you only see it in negs?
You see, if the mask is not working "optimally", once you compensate it out, you can actually have "negative densities" or "blacker than black" if inverted to positive, due to imperfect masking. This would be clear to eyes when looking at the negative, but would clip to black in any printing or scanning process, analog or digital. Or, to rephrase; if the mask reacts without other imaging taking place, it will create a weak positive image, so, it cannot appear in print blacks because it will be even more black. Maybe this is the case? If you use a scanner to "lock" on the Dmin, this kind of light-fogged piece might then be problematic for this calibration; then, after locking to these "blacker than black" densities, normal densities could look like colored fog. But if you print it with your normal filtration and exposure, it should look normal.
Just a thought.