My take on it is that tighter tolerances (smaller, unspaced coils) tend to hold high heat and liquid together in close proximity, in effect "cooking" it, in addition to providing a lot of nooks and crannies for build-up to start. What you build on has a lot to do with it too, I think. Tightly enclosed build decks with narrow chimneys and limited airflow, such as on a Kayfun, basically act like little ovens once the airflow stops. Compare that to your garden variety RDA, like the Velocity clone in my hand at the moment, where I can look down into it and see the coils, so it's not dependent on vacuum (ie: drawing on it) for airflow. Just sitting there, it's being cooled by air currents in the room, which isn't possible with coils that are buried inside tanks like a Kayfun.
So while I think some of it has to do with the gunk potential of a given juice, it also has to do with the style of atty, mainly how efficiently it dissipates heat, and coil build. Heat, heat, heat. Have you ever seen a dual coil build that's been firing unevenly? The hot coil is the one getting all the build up.
At this point I have really limited experience with NETs, but I'm finding that they seem to perform better in RDAs than tanks, and for some odd illogical reason, with Clapton builds. The "Black Cavendish" stuff I had (and loved) pretty much wrecked havoc on the 24 g 2.5 id unspaced dual build in my Aromamizer, which has a REALLY tight, enclosed build deck. 5 mls of the Cav and it was shot, requiring a toothbrush scrub to remove tarry residue and full rebuild. For some reason the dual 2.5id spaced Claptons (0.3 ohms) in my Griffin fared better, and my Velocity RDA (build identical to the Griffin except 3mil id) just needed a quick dry burn/rewick. Strange, I'd have thought Clapton wire would gunk quicker, but it's the opposite (thicker wire, spaced = better heat dissipation??).
So, try dual builds on an RDA or Griffin-esque tank, and experiment with Clapton wire.