Please remember that (from experience) you are more of a compiler while I confirm, which can result in different confidence thresholds and different weights for evidence. This means I may note and weigh things differently.
For instance, "I Vår Vackra Vintervärld" has a 1995 date in the autogen, so I indexed the recording from "Stjärnan Lyser Klar" [2000] to "Julnatt" [1995]:
Provided to YouTube by WM Sweden
I vår vackra vintervärld · Sten Nilsson · Joybells
Stjärnan lyser klar
℗ 1995 Mariann a Warner Music Group Company
Miscellaneous: Peter Himmelstrand
Composer: Felix Bernhard
Lyricist: Lasse Green
Lyricist: Richard Smith
Again this contrasts with "Låt Julen Förkunna" found on "Stjärnan Lyser Klar" [2000]:
Provided to YouTube by WM Sweden
Låt julen förkunna · Sten Nilsson
Stjärnan lyser klar
℗ 2000 Mariann a Warner Music Group Company
Lyricist: Dan Hylander
Writer: John Lennon
Lyricist: Py Bäckman
Writer: Yoko Ono
In this case I don't have even have a non-autogen indicated as from "Julnatt" [1995] with which to compare audio or runtimes.
When I know that I have a later version of an unconfirmed prior version, I will add a date to differentiate: Nilsson, Sten - Happy Xmas (War Is Over) [Låt Julen Förkunna] [Swedish] [2000]
In this case I added no date but would have known about the track listed on "Julnatt" [1995], which suggests that I left the question open.
This all means I see nothing "surely" about this question. Wouldn't really matter for SHS, as re-recordings are not added, Management is hostile to even noting re-recordings in the comments, and multi-indexing is discouraged as well.