Soundstage, layering and tube differences

Your observation about SED's vs Mullards IMHO falls right into the same observations that can be made about equipment in general. Stuff (including tubes) that have a 'full bottom end' more often than not obscures some of the information present in the high frequencies. SED's have a tendency to be a bit forward in the upper mids but are not bottom heavy so you can get better dimensionality when using them. The Mullards probably are a bit 'warmer' than the SED's therefor the differences in imaging that you have noticed. Of the three new(er) EL34 production type tubes I enjoy more often than not would be the JJ E34L or NOS Tesla E34L (their predecessor).

FWIW, a tube I often use in lieu of a SED EL34 is the SED 6l6gc. A very linear tube (if that suits your needs) emphasizing neither the bass nor upper mid range emphasis of the other tubes you are using. 

The bad news - SED's have been discontinued as a new production tube.

Welcome to tube rolling! Yes, this is normal and indicative of differing tube quality and methods of construction. +1 for 6L6GTB can be sweeter and clearer than EL34. If you have some coin, tbere are some GEC KT66 available.
https://www.tubedepot.com/products/kt66-gec
BTW how did your termite tenting work out?

Oh...I've been tube rolling for many years, but I've never had my question answered by someone who really knows. Is it the actual recorded soundstaging that some tubes resolve better or... just an illusion form the harmonic distortion of that particular tube?

jim, If you don't know the answer to your question, how will you know that anyone who answers really knows? Go figure.

Anyway, the 'illusion from the harmonic distortion' is related to the difference between solid state devices which typically have odd order distortions and tube devices which have even order distortions. Some feel that even order distortions are additive and odd order distortions are subtractive. You can easily research this issue - it is talked about quite a bit.

Individual tubes differ most audibly in tonal balance (but not exclusively), as do most of the other components in your system. Some are subtractive from the original source but none can be truly additive even if you like the sound better. The pits and grooves are absolute - its what you do with or to them that counts. 

My theory about soundstage is if the musical recording does indeed include the soundstage  "information", then any component in the audio chain - tubes,  cables, source gear, amp, preamp, etc., etc., as well as room acoustics  and speaker placement, can and will contribute to rendering that information. If everything is the same and a change in the tube brand renders a more realistic soundstage, then one could surmise that it's the tube that is delivering more of that information. Don't know if this qualifies as an answer but it's my opinion.