It is an interesting article but misses a few points. One is that the harder brass is, the faster it anneals. 90% hard brass exposed to the same annealing temperature as 50% hard brass will recrystallize nearly ten times faster. Fortunately, brass's modulus of elasticity is constant for all hardness levels and degrees of annealing. So while we don't want to under or over-anneal, there is a surprisingly large range of annealing completeness over which we will get the same apparent performance as long as the interference fit between the resized neck and the bullet being seated is the same. Note that because harder brass is more springy than anneal brass, this has to be checked.
Second, recrystallization and grain growth you can see on etched brass with an optical microscope, are not the only stages of annealing. Before the brass heat exposure reaches those stages, it undergoes
recovery, which relieves stresses on an atomic level, and which restores some malleability and is thus enough to prevent neck splits. Recovery, being at the atomic level, is not visible in an optical microscope and has to be "seen" by x-ray diffraction.
To achieve recovery to prevent neck splits, the lower heat of the candle flame annealing method or the molten lead method mentioned in the article can be made to work. Both are slow, but as described by metallurgist Fred Barker, they are:
(1) Lead Pot Method: heat lead to 725°-750°F; dip neck into powdered graphite and then holding body of case in fingertips into molten lead: when case body becomes too hot to hold slap case into wet towel; or
(2) Candle-flame method: Hold case body in fingertips, place case neck in flame and twirl case back & forth until case body is too hot to hold, then slap case into wet towel; wipe soot off neck & shoulder with dry paper towel or 0000 steel wool.
Fred Barker, Precision Shooting Magazine (RIP), July 1996, pp. 90-92
The wet towel, in these instances, is to stop heat from spreading toward the head and help remove candle soot and lead oxide, and flux traces. With the candle method, I have not found it necessary, and the soot can be removed later. However, the candle method is painfully slow, as you are often about 20 seconds per case.
My point in mentioning these methods is to illustrate you need less annealing than you might think, and if you do go to recrystallization, you are doing more than is minimally necessary but aren't hurting anything if you take it that far. You won't, as a practical matter, see any difference in neck tension over that range, provided the necks are coming out of the die the same diameter. If a case gets too soft, its yield point (where it goes from elastic to plastic deformation by an applied force) gets too low, and there can be too little spring left in the metal to hold onto the bullet well, especially if you anneal after resizing and don't reinstate a little work hardening by sizing.
The above range is a good thing because of the recrystallization time difference I mentioned before. Even with identical heat exposure, those with brass that hasn't been annealed for a while are taking their annealing further than it goes when the brass has only been reloaded once.