YG, Magico, Wilson included, I have heard no improvement from anyone allegedly making complex crossovers an ameliorated driver of musical destruction. In those cases, actually, quite the opposite. I wasn't, by the way, a single-driver enthusiast from Day 1. It was an ideal unrealized. There were no authentic sounding full range drivers throughout the 70s, 80s and 90s. In the 2000s we got some breakthroughs, Zu chief among them and as you point out, no Zu speaker is a true single driver transducer. Most Zus use various grades of their continually-evolving FRD covering 90% of the musically-relevant octaves range, complemented by a supertweeter on a high-pass filter for harmonic completeness, and then two models thus far have been further complemented by amplified sub-bass modules on low-pass filters. But the result is Zu gets closest to full range, and does so sans crossovers, and delivers high efficiency with high power handling. They are friendly to a very wide range of amplifiers.
As I hear them, modern multi-way, multi-driver, crossover-intensive speakers are losing ground in coherence in part because the resolution of the individual drivers is improving faster than the deleterious effects of using them can be ameliorated. YG & Magico, just as examples, while sounding very different nevertheless make the impression their designers never actually listened to their designs before implementing for production. Do they transduce signal into sound? Absolutely. Do they transduce signal into music? Alas, not so much. Most of today's designers are going forward on linearity and backward on coherence and tone, with crossovers introducing dynamic non-linearities to further pollute the result. Amps have for the most part gotten better, even the bad ones, but the speaker makers have mostly lost the plot. Or listeners have, so they reward designers' wayward behaviors.
I posit that if the aggregate design talent currently misdirected and wasted on dead-end multi-way speakers instead followed Sean Casey's lead and put their minds to crossoverless design, we'd have a cornucopia of musically-convincing loudspeakers, faster-paced evolution of full-range drivers, and a conclusive endpoint shoving the crossover-based speaker aside to irrelevance. It's not that people aren't smart enough in this endeavor. They just lack imagination and will. The development path for multi-way speakers leads to a dead zone. The soil is depleted. Ideas have dried up. We're converging on the same, flawed sound. The development potential for the crossoverless speaker is however vast, but it takes work. What Zu, Voxativ and Audience have achieved over the last 15 years in advancing coherence in loudspeakers is just a start, whereas everything invested in multi-way has been a finish. Multi-way sound isn't much better today than in 2004. Just more ways to hear the same errors.
Why aren't there more audiophiles? Because complication and cost are rising, but musicality is not. And speaker makers are the leading cause.
Phil