Hm it looks like there won't be a straight upgrade path from Leap 15.6 (SLE 15-based) to Leap 16.0 (SLE ALP-based). So I don't think anyone would dare to make a new Leap installation.
But maybe there's no real need for Leap anymore. For desktop users, Tumbleweed is the best choice. Thanks to Snapper and OpenQA it's hard to break, it basically only happens if third-party repos like Packman or Nvidia are out of sync but then you can always roll back and wait a few days.
I think it would make sense to have new SLE clone but then it should also benefit from the same support timeline. Leap minor versions are only supported for like 1.5 years and you have to upgrade between the minor version using the CLI. SLE calls the minor versions "service packs" and you can do the upgrade with a couple of clicks in Yast. Furthermore, SLE 15 has 10+3 years of support which - like with RHEL - would be the main reason for anyone to use it. I don't see the point in using old packages including an ancient kernel with random backports like in Leap but without the long support horizon that SLE has. And to make it weirder, you have the newest Plasma and Xfce versions because those are NOT coming from SLE binaries but from Tumbleweed. It's all a bit schizophrenic. Currently Leap is neither fish nor flesh.
So after Leap 15.6 dies, I'd rather see:
1. oS Tumbleweed, reliable rolling release with the Plasma, Gnome and Xfce desktops officially supported. Upstream to SLE. Similar role that Fedora has.
2. oS Aeon/Kalpa, an immutable version of Tumbleweed, currently only with Gnome and Plasma versions. Compare with Fedora Silverblue/Kinoite.
3. oS "New Leap", a free version of SLE ALP, focused on servers and such. Same decade-long support horizon. Compare with Alma Linux vs RHEL.