Will Tipton's book chapter on indifference and the breakdown of indifference in the case of asymmetric ranges is killing me as I find myself returning to it again and again. My most recent confusion revolves around the minimum defense frequency (in very simplified terms MDF = 1-A but of course defending vs. bets is more complex than that) and which part of the bettor's bluffing range the MDF actually makes indifferent between betting and checking?

Background:
The MDF dictates which range portion the defender needs to continue vs. a bet in order to neutralize the bettor's incentive to add any further bluff to his betting range. If the defender reaches the MDF, the bettor cannot profitably expand his bluffing range as these will break even long-term. If the defender fails to reach the MDF, the bettor is incentivized to bet any two cards.

Question:
Specifically I have the following question: When the defender defends exactly the MDF, is every bluff within the bettor's betting range 0EV and hence indifferent between betting and checking? Here, indifference would occur across the whole bluffing region of the bettor. Or is only the worst bluffing hand/best checking hand indifferent between betting and checking? Here indifference would only occur at the threshold between betting and checking. I have the feeling that I have heard both reasonings already, e.g. in Steve Paul's video on 1-A here (example for first reasoning) or Krzysztof Slaski's recent video here (example for second reasoning).

Attempt of an answer:
My personal take on this would be that both logics are correct and it depends on the situation when which logic applies. I'm not sure about this though.

Spot 1: River bet
On the river, it makes complete sense to me that all bluffs are 0EV/indifferent between betting and checking. Imagine the game of nuts/air vs. bluffcatchers. Here, the MDF is defined exactly by 1-A. All of our bluffs will always lose if we bet and get called or check no matter their raw hand strength (two overcards vs. 6 high) as they by definition lose against all bluffcatchers. Hence as the defender reaches the MDF, we are indifferent between betting and checking with each and every bluff.

Spot 2: Flop bet
However, shouldn't this be different when we bet on an earlier street and our bluffs still have backdoor equity? E.g. does reaching the MDF (whatever the concrete value might be OTF, somewhere around 1-A but this depends on the spot) on a board like Th8h2s really manage to make a hand like AhKh indifferent between betting and checking? I can almost not imagine that as it seems to me that this type of hand actually has an intrinsic incentive to build the pot. Here I guess indifference only occurs at the threshold. If the defender continues enough of his range, the bettor cannot profitably expand his bluffing range since the defender responds appropriately. However, that does not mean that a hand like AhKh is not a more profitable bet than check. In this example, indifference would hence not occur across the whole bluffing region but only for the worst bluff/best check.

So I think that the reasoning of "all bluffs being indifferent between betting/checking" breaks down in a multi-street game since many of our "bluffs" can hardly be classified as such, the distinction of value and bluff-bets actually only really applies to the river, right? Digging even deeper, would it even make sense to classify a hand like AhKh as a bluff on the Th8h2s board? It seems to me that it's just a very human tendency to dichotomize our range in two easy parts. I would say AhKh just has an overall incentive to move money into the pot due to its equity here, I don't really care if I'm bluffing, value betting or something in between that we could define with a fancy new name.

Am I going wrong?