There are many more expensive choices to be made in a crossover network, and I have no question that it can make a difference in the sound.
The first "impressive" network that I owned was some time back, in my JBL 240Tis. They were an impressive network with many capacitors, most of them film, and a couple of EL caps. This network did some shaping as well as simple band-pass/high-pass/low-pass and was in my opinion not built to any requirement to save money. The EL caps were there (in my opinion) for either of two reasons: 1) They didn't make any difference to the SQ as they were simply clamping and not directly in the signal path or, 2) they did make a difference in SQ and the somewhat "slower" nature of the EL caps was preferred.
If it was the first reason, then no harm in changing the couple of EL caps out to film. If the second then you might be changing something that introduces a characteristic that the Designer (who we all know, and who has written that networks were his passion) didn't want.
These speakers, like all JBL speakers, were designed "on paper" by a well-respected Designer, prototyped and auditioned until they were as desired, then launched into production without changes to values nor types of passive components (it would be highly unusual for anything to change from the approved design at that point). Whether the addition of Monster Cable interconnects was for marketing or for real SQ I might debate, but that's another subject entirely.
My point is that they were as good as they could be with the original components, as were many other quality speakers (in and out of the Lansing/Altec/JBL family) and although there are certainly improvements that can be made with our 21st century technology, I have a hard time believing that 20th century changes like swapping in film caps are really making them better. Different perhaps, but better? If I still had them I might consider some cabinet bracing however.
That having been said, and this already a long post/rant, I also feel that it is important to remember that at this point for most of us this is a hobby and all is subjective. What sounds good sounds good. I've been guilty of building my own computers, engines, drivelines and suspensions, audio components, etc., etc. and making modifications that feel good or sound good.
Jbl and everyone else made speakers for the masses, tested them in their multi-million dollar anechoic chamber and facilities with expensive equipment that we don't have. Their goal was to sell excellent speakers that will sit in the average listener's living room, hooked to the average person's system, playing the average person's music choice, to the average pair of ears. That is where the money is.
What sounds good to me, or you, or anyone else in our particular environment can be completely different and if charge-coupling with 9VDC, film-caps, air-core inductors, silver cables on stands etc. sounds good in your room then that's all cool. It isn't a crime to customize or modify something to better suit your needs and it's one of the most fun parts of this hobby.
If there isn't any measured data showing improved imaging from the charge-coupled crossover it lacks scientific proof, but apparently many people still are happy with the change.
I'm in the active camp anyway. All of that reactance in passives, phase-distortion, inability to tweak and make adjustments doesn't feed the obsessive side of me and I need to feed the beast.