I won't be using 1804a for a few reasons: the nanocons are really small and I don't think it would easily fit. Also, there is no advantage to star quad for a cable run that isn't balanced because there is no common-mode rejection. It would actually be worse because the capacitance of star quad is higher than regular cable, and the connection between coupling and PFA is still relatively high-impedance--I can lower the impedance (that is actually set by the PFA/amp), but that will lower gain. At higher gain, 10ft would be the limit; lower gains could probably support up to 30ft although it's still an unbalanced connection, so interference can also start to become a problem. That is with Mogami 2697.
I could avoid those problems with a 4-wire connection, but that would exclude the nanocon. I would have to make the coupling larger to support a 4-pin mini-XLR; then I would use 1804a, but not wired as star quad.
I'm not sure it is possible to move away from the nanocons (may already be in the process of getting the collettes milled or whatever), but if it isn't too late in the design phase, I'd say it is more important to get a good circuit developed and provide better performance. It sounds from what you're saying that a 4-wire operation would yield better performance, or at least longer unbalanced cable length.
For my own use with "active" cables, I sometimes use them for low profile recording, and in these cases 10ft length is fine. I also use the active cables for regular taper section recording, and in this case 10ft is about the minimum length. That is the length of my milab actives (3m), and I often wish I had just a couple more feet available (esp if I am clamping onto someone else's stand and can't get my bag right under the stand).
Not that any of this can't be worked around, but the nanocons sound like they limit performance somewhat, and they are pretty thin -- which sounds like they might be more fragile and you've stated it probably wouldn't be good to use them to connect to a mic t-bar. All that said, the Binder 711's seem nice too. If you consider the total length of cable-end connector plus active-head connector, they aren't much difference in length than the nanocon combo, and I don't think diameter is much of an issue. A trade-off perhaps in outer connector length vs connector length inside the active head, but either way I think the total length is comparable.
The Binder connectors might also be more conducive to using in a kwon-bar (delrin rod) type set up for connecting and shockmounting the caps.