In addition to moster67's suggestion, NetBeans also has a NetBeans Platform upon which you can build great apps. And this ties into my second point: making the B4X IDEs with B4J would make B4J better. In this case, by adding a new type of B4J project: NetBeans (or IntelliJ) Platform Project. Lacking that, I think the RichviewFX library looks promising:
https://www.b4x.com/android/forum/threads/custom-view-richviewfx-codearea.62823/#content . Either way, there are lots of world class IDEs that run in the JVM.
To JakeBullet70's question, I'm not sure that that statistic is that important. Obviously almost all B4X users will have at least one Windows system since B4X is primarily Windows-only. But how many potential B4X users are out there who don't use Windows? I suspect it's a large untapped market. This relates to a point I've made in other threads about evolving B4X in ways that don't
only appeal to former VB6 developers. Furthermore, how many users are tied to Windows solely because of B4X? That latter case is certainly my situation and I know I've seen that opinion posed elsewhere in the forums.
As to Erel's point about the size and difficultly of such an undertaking, I think Anywhere Software could recoup the cost of such a project (and then some) by keeping and advancing the Windows-only versions with their current pricing model while also offering cross-platform versions to be purchased separately. At least until the cross-platform versions achieve feature parity with the Windows-only versions at which point I assume the Windows versions would be abandoned. I would happily pay $20/IDE/subscription period in addition to the Windows-only versions' cost to be able to ditch Windows part-time in anticipation of feature-parity cross-platform versions that would allow me to ditch Windows full-time. And I'm sure I'm not alone in this.