I mentioned it earlier and thought I'd expand on that.
I first tried it as per directions and got OK accuracy, enough for hunting. But not as good as other bullet coatings. Was talking with Dyotat100 and he said he'd tried it but got poor accuracy because it gave a wide ES.
So I looked at what was going on and thought it might be because the coating was too thick. The next batch I tried I used more solvent and made a thinner coat and the accuracy got a bit better. I was still doing two coats so I dropped to one and the accuracy improved more.
I shoot with the rifle clamped in a rest where it can't move at all, all that is touched by me is the trigger to take the human element out if it and only test the rifle and bullet. I did not try coating swaged bullets like the NSAs, but the cast bullets I used for these tests were inspected and weighed and were as close to perfect as I could get (cast over 400 bullets to get 50 for the test). After coating, I sized some to .357 and some to .358. the rifle I used is a Cobra regulated FLEX that shoots the 147gr bullet at 920 FPS for over 260 FPE. It is a very accurate 1/2 MOA rifle. The barrel is a 24" Green Mountain 20" twist that has a collet tension system instead of the circlip usually used and so can be tensioned much more than usually seen in airguns (20 foot-pounds).
With only one thin coat, even though I used Zombie Green you can barely see any color, but it's was keeping the fouling down and the selected bullets gave the accuracy the gun is capable of. I had coated a bunch of the not 'perfect' bullets too and shot over 200 of them and although there was some slight fouling that I think might have been traces of Hi-Tek, I did not see any leading.
Note that this particular bullet casts at .359 so it's not getting a lot of sizing and the thin coat stayed on the bullet just fine. I have not tested it on a bullet that gets a .003 or .004 reduction. I did use the thin one coat on some .25 bullets that sized from .254 to .251 and they kept their coating and shot well.