I think you got thing mixed up here, the loop on reply #83 are the same with your (Double Girth hitch Eskimo bowline), the difference is left and right hand version only.
Noope !

Just follow the Standing Part, after it enters into the knot s nub : How many turns does it make, before it forms the rim of the "upper" collar ( the collar around the standing end ) ?
Two, in the case of "your" loop -
One, in the case of "my" loop. Mirror symmetry does not affects
numbers !
Your new loop, based on a Pretzel-like nipping structure, is fine, indeed - and those two crossed round turns keep the eye leg of the standing part aligned at the centre of the "lower" collar and of the whole nub - a more balanced arrangement than the one you had used in your previous loop. I, too, had used this nipping structure a number of times, and I have seen that it is very tight and stable - but I have not been able to persuade anybody about this !

See how efficient it is, its superb gripping power and its great inherent stability, even in the case of a simple adjustable "Eskimo"-like loop, where there is no collar structure at all.
However, in the case of a double nipping loop / double collar bowline, I believe that the Pretzel-based nipping structure does not offer anything more than what the simple, much easier to tie and inspect Girth hitch does. Also, the Pretzel has those two "odd" X s = two crossings, one at each side, which disturb the smooth, nice flow of the lines we see within and around the nub of the Girth - hitch based loop. So, although I, too, was seduced by the Pretzel nipping structure ( when used in "common" as well as in "Eskimo"- like eyeknots ), now I have "returned" to the more simple Girth hitch.... Anyway, both loops are almost equivalent in every aspect, and my recent preference for the Girth hitch-based one might be nothing more than a matter of personal prejudices or taste.
De gustibus et coloribus non est disputandum... I do like your laboratory, very much ! I hope that I will persuade you to proceed, and perform some destructive tests some day, so we can have an idea about the ultimate strength of all those loops.