Mark and Don, thanks again for your interest and many comments. I have some new findings and progress to report. To recap, I had an on-going problem with overinking on Epson Enhanced Matte paper--
>I see mottled or "water-stain" artefacts in EEM/4800/MK most easily under strong incident light at a 45 degree angle. Previously I had said that this effect was restricted to the darkest of dark tones, but now looking more carefully at my entire test chart (see attachment in previous post), I can see significant reflective mottling in several of the Gretag Macbeth Color Chart patches (e.g. moderate red, purple, red, neutral 3.5), and less significant but still clearly visible mottling in several other patches-- again, when the print is held at an angle.
I thought that a good solution might be to switch to a heavier, all-cotton paper: ink would absorb better, and I might even get wider gamut / improved Dmax. Unfortunately, Hahnemuhle papers (which seemed a consensus premium favorite) are too rough / not bright enough. Based on recommendations to try one of the Red River papers, I got samples of their Aurora bright white, as well as the GC (greeting card) paper. Unfortunately, the Aurora has too much texture for my taste, and the GC is non-archival.
So it looks like I'm "stuck" with EEM for the foreseeable future, and it's up to me now to solve the overinking problem. Epson "pro" support was no real help, unfortunately. The guy knew that overinking (which they call "caukelling") can happen, and all he could suggest was to adjust the color control params in the driver, or switch to their Ultrasmooth Fine Art paper. Unfortunately, there are just too many variables in the color control dialog to optimize over; and UFA paper isn't bright enough for me.
I've corresponded with Andrew Rodney about some of my problems, and he's suggested considering a RIP in order to have finer control over inking. Because this would be quite an expensive step, I'm checking out the docs carefully. Interestingly, I found that my "water-staining" problem appears to be described in ColorBurst's documentation as "inkjet reversal"--
When the channel ink limits are not set correctly for a particular media, a wide range of problems can occur. If the ink limits are too low, your prints may show banding or you may end up with low ink densities. If your ink limits are too high, your prints may be too wet or you may experience inkjet reversal. Inkjet reversal is a phenomenon that occurs when an optimal ink limit has been exceeded—as more ink is laid down on the media, less color saturation and/or density is achieved.http://www.colorburstrip.com/windows/Spect..._Win_Layout.pdf
I would feel more confident pursuing this RIP path if I knew chances were high that a fully optimized RIP print would be noticeably "superior" in some visible way over a fully optimized "Epson driver" print from the same system.
Any comments appreciated,