I am not sure that I mean either, but probably closer to the latter. For example, I could do 5 x1000m at 3:07 with 3:00 active rest as kind of the bare minimum to get a vo2max adaptation. But if given my total running volume and my fitness 8 x 1000 at 3:07 with 2:30 rest is a workout that I am capable of and is consistent with my overall training volume and recent workouts, I would certainly prefer to do the 8 x 1000 workout with 2:30 rest instead of the 5 x 1000 workout with 3:00 rest even though 5 x 1000 with 3:00 is the minimum necessary to get a meaningful vo2max adaptation.
But I would not advocate 5 x 1000 at 2:58 with 1:30 rest. There would presumably be more time at my max heart rate with that workout (and thus more vo2max stimulus), but the additional time would probably be marginal (especially for me since I am running in the heat and I am middle aged and both of those result in my heart rate climbing back up pretty quickly when I start in interval and thus offsets the need for faster intervals and shorter rest to get greater time at my max hear rate). The increased stress does not yield enough of a meaningful benefit to warrant it from a Daniels perspective.
As to the other benefits from such a workout beyond vo2max, I absolutely recognize that many people attribute value to those benefits even though they are not well-identified by scientific studies. I also get that missing out on whatever benefits that may come with those in-between paces is probably the biggest or second biggest criticism of Daniels. But I think that once you start trying to seek those benefits, you are no longer obeying one of the cornerstone principles of a Daniels approach. That doesn't mean it is not worth doing, it just means that you are no longer training in a Daniels model (which goes to one of the primary questions that I was trying to raise with this thread - at what point does a change mean you are no longer training per a particular coach's principles?).