A correction. I'm posting this all without my notes, which may not even exist anymore. It should be obvious that you can't project through an orange mask and get an orange mask on the RA4 interneg. So, you have to project through the opposite. The mask should consist of 60 cyan and 20 magenta. This will then give an orange mask on the interneg. I think that is right. I'll continue to look for my notes.
polyglot: Yes, if you didn't have the positive to positive chemicals and papers, and didn't want a print made using digital, the likely best way to go would be to make a C41 internegative. But to do so, you have to copy the transparency onto the film, process the film, and then enlarge the film. I was looking for a way to quickly do a positive to positive and remain in the RA4 workflow. And after all that I've done, the C41 route is the only way to go if you need quality.