Why is one el84 running much hotter in my Blues Jr.

[pdf64]

It actually adds something like 10pF across the primary. But this just adds to the 100pF to 500pF primary self-capacitance of the OT. These are numbers from real measurements.
So its effect on upper frequency limit and phase lag should be negligible.

As with twisted plate wires the electric field is confined between the wires there is much less unwanted coupling to other circuit points.

Yes, my understanding is that it's good practice to twist wiring carrying a balanced signal, it's something I've always done, even if it's only a turn or 2 along a short run. Obviously the longer the run / bigger the signal, the greater the benefit.
I guess that it takes a few moments to do, so manufacturers don't bother.
And because manufacturers don't bother, neither do copycats or DIYers.

[Helmholtz]

Twisting of balanced (opposite phase) signal wires is quite effective with input (interference-sensitive) as well output (interference producing) signals. Of course standard shielding (with ground reference) would do as well.