I did worry a bit that my suggestion would require processing you don't have. I kind of feel like we all could use a bit more audio processing (of the good kind) in our lives. I am an engineer, and I designed my own custom processing because I couldn't buy what I wanted.
A 5.1 psytrance album? That sounds like something I'd love to hear. Can I buy it somewhere, on a DVD or as a download?
Well first, I wouldn't really do anything different for psytrance vs. other genres. My specific focus would be on sound quality over quantity, call it semi-pro if you like. I'm thinking it'd do well for small churches, small cinemas, etc. in addition to stuff like you're doing. To be honest, my idea is far from fleshed out, in part because there are a lot of mundane considerations to settle: size, form factor, HF driver height, etc.. Lacking real world live sound experience means I don't have a good grasp of best practices for that kind of stuff. What I do have is strong theoretical knowledge and experience designing and building the system in my living room along with a kind of working hypothesis about what happens in larger rooms and outdoors.
For mid and high frequencies, I am a big fan of the SEOS waveguide design, which provides highly consistent frequency response at multiple angles. This is a beneficial property in small rooms because it makes the reflections more similar to the direct sound. It seems logical that this property would also be excellent for covering a wide listening area like a dance floor. The pattern is nominally 90 degrees (i.e., -6 dB @ +/- 45 degrees), but sound really doesn't change much at steeper angles, it just kind of gets quieter. To my knowledge, the SEOS waveguide has not been used in a pro speaker design. This is probably because the waveguide is physically large relative to their low frequency limit and output capability. I don't think that would be a problem in this kind of application though.
My home speakers use the 15" wide SEOS, but for this purpose, I would probably go for the 24" wide. Forum member @lukeamdman used the 24" for his speakers. I'd probably use the same coaxial compression driver, the BMS 4594ND which contains two drivers in one package and extends up to and beyond 20 kHz. I'd probably cross it a bit higher than he did, at like 600-650 Hz, but the details would depend on what I used for the lower part and what the measurements looked like after I got the equipment. As for the lower part, I'm not sure what I'd do. I'd probably look at horn options for the bottom, but I don't know enough about horn design to know whether it's practical to cover a wide low frequency range, like e.g. 80 to 800 Hz without it being too large. I don't think a horn with folds would work. I would have to research it.
One possibility that puts the HF driver at ear level (maybe much too low?) is to use a 2x2 array of 12" drivers, probably in sealed cabinets. The array would be adjacent to the ground and would couple well with it. The 4 x 12"s would have lots of headroom too. A single 18" or 21" would probably save some money, but not many drivers that big play well up to the c rossover. And furthermore, I think I'd want more headroom than a single 18" offers, which complicates the design in ways I don't like. As for whether to do sealed or reflex, it would really depend on the driver and what my low frequency needs were. Sealed is smaller and simpler and more likely to work with Skhorns that go higher. Reflex is larger but can play lower to mate with Othorns. It's not that simple though. I'd really need to look at different drivers, do simulations, etc. to reach a decision.
Finally there is the processing. At the least, some high-shelf EQ is needed to correct the downward tilt in the treble which naturally occurs, in addition to crossover filters. The SEOS doesn't need as much boost to the top as some horns do though. In my case, I have my own processing and I have been working on novel methods for speaker sound quality optimization. I would use that to get the best possible sound out of the whole system.
I don't actually have all this processing capability yet. My home system is the prototype, which I'm still working on perfecting before I move to testing on other systems and rooms. Like I said this project is basically fantasy right now, but if it were to happen there's a chance of me getting to it in a couple years.