Here is a thought, or series of them...
What does one need in a color film? For me, it is the ability to capture the color information of a scene, in one exposure, in perfect registration and in any size, the larger the better. The size caveat is important in dismissing digital...
For me, I have less interest in printing this picture on RA-4 or Ilfochrome paper, because I'm operating under the assumption that if I'm making a color emulsion, none of these great things will exist anymore, let alone color film, so why count on any paper being around? Carbon, carbro and dye-imbition (DCG) are what I'm hanging my hat on for the future, as the materials are for the most part independent of silver photography. And of course, scanning.
So, with this in mind, the color emulsion need only have the color information stored in it, not necessarily the color appearance stored in it. What I'm getting at is, it doesn't matter what the color emulsion looks like, as long as the 3 color sensations are recorded and are separable by use of appropriate filters.
Point being, all color films have sought to reproduce the scene as realistically as possible, since they were either slides or designed as negatives to print on a similar material coated on paper. But had someone been making a film designed for one of the separation methods, or today, where digital capture and "decoding" is an option, the film image need not resemble the scene in color.
I don't know enough about separation filters, efficiency thereof, overlap, etc., to say if such a scheme could work; but if one is making a color emulsion and their stumbling block is the appearance of the dyes, it might not matter afterall. As long as they are different enough to be discernible in some way.
Correct me if I'm wrong, but Gasparcolor utilized a scheme where, as a printing film, it could not be utilized for direct capture. Instead, separations were printed onto it using blue, red(?) and infrared light to reach the sensitive layers. This is of course entirely different from what I just proposed, but it's an example of "thinking outside the box" in terms of color emulsions.
This is all theoretical of course, but I hope what I'm getting at makes sense to some degree.