I finished reading the RAIS thread a couple of weeks ago, very interesting and a methodology I would aspire to - I have got as far as a GDS front with a coupler so the pseudo-fin or intake part can be changed over on the same motor mount (to trial different shape/area inlets as well as making it easier to change out toasted ones), plus then an open fronted cylinder with some slot on rings with different internal and external fin arrangements...
There is a slowly developing forward facing cone idea in my building box which would be GDS-ish but because of the forward facing-ness it I think it could also have a RAIS-like contribution. In the same vein as one of the posts on your original thread about a GDS-RAIS hybrid I suppose. I have a nice light plastic cone which is only mildly tapered to give this some chance, will update as and when I get anywhere with it! Also keen to see outcomes of other peoples (careful) experiments along these lines.
The key issue with GDS that I think I have come up against a bit is that it wants the motor-exhaust velocity to be high compared to the model-through-air velocity, preferably a lot higher I think, esp if the stability is already marginal. This doesn't work that well with typical BP motors because they all have big take off thrust spikes. I'm trying to get on to low impulse long burn APCP (Klima) motors (starting with B2) to get past this with one or two small/simple models, like the simple
Black Arrow I have made and started to trial, fingers crossed....
My estimation is that a forward facing cone may need some very strong ducting/induction stabilisation however this is done because not only is it unstable CP-wise but is also has a strongly positive delta (frontal SA offset)/delta (theta) characteristic for small angles of attack, ie the cone will always increase its veer and spin with even a tiny initial offset (positive feedback), whereas the conventional point-forwards cone (ie a nosecone!) has the opposite characteristic where it will tend to self-correct for small aberrations.
The other part that I think may be difficult for a cone is getting the change in momentum of the air (whether ducted or inducted) to be far enough from the CG to get a decent moment, without which I think the stabilising effect is lost.