As far as the "temporally stretched formula layers" are concerned, remember that Stockhausen does this on many different time scales. If there is an "experimental" dimension to formula technique, it is in the attempt to see how far human perception can be stretched beyond the boundaries usually regarded as absolute limits. It is commonplace for composers to present material in augmentation, where a theme is doubled or quadrupled in length from its "normal" speed, and of course Stockhausen sometimes does this, too. But how far can he go before the whole character of the sound is transformed into something new, like the famous transition from pitch into rhythm in Kontakte? Further, can human hearing learn to recognise the relationship between such radically transformed states of the same material, and identify them as being "the same"? Stockhausen was always fascinated by these boundary phenomena, as well as the idea that things on drastically different scales of magnitude might be objectively "the same", and yet be perceived as entirely different things.
On a fairly modest level, when I first heard the Electronic Music with Tone Scenes from Freitag as Licht, I was fascinated by those sparse bass tones that set in after ten minutes or so. It quickly became obvious that they occurred at regular intervals (something under a minute apart, I think) but why did each new pitch seem like just the right note? Indeed, why was I able to anticipate the next pitch before it had yet sounded? The notes in question are of course the "scale" figure in the third bar of the Lucifer formula, and they seemed so logical and inevitable to me because I already knew that twelve-note rising sequence very well from previous incarnations. But it still came as something of a shock tireless I was recognizing a familiar melody stretched to something like 150 times its "normal" length. (I think the clue that tipped me off was the trumpet-like figure that occurs a few minutes later, which is much more easily recognisable as Eve's "Wednesday" segment, where the opening grace-note figure now has the notes changing every ten or fifteen seconds.) This set me to wondering if even longer stretchings could still be heard in this way, and I have no doubt that Stockhausen must have thought along very similar lines.
I like that analogy to the well-designed building, too. You don't have to be holding the plans in front of you to sense whether certain relationships between rooms and hallways are sensible or not—at least, up to a point. As you travel through the building many times, you gain confidence that you know where things are, and anxiety about getting lost subsides. The same is true of a countryside that becomes more familiar the longer you tramp around it. At a certain point, you attain an overall mental picture of where everything is—or, at least, how various prominent features stand in relation to each other. The sense of scale, however, may be distorted. I think we all have had the experience of walking or driving over a once-familiar but not recently travelled route, and repeatedly thinking something like, "that church with the odd steeple is right around the next bend", only to find that there is still more road, and in fact four or five more bends in it before the familiar sight comes into view. I think that larger stretches of music are often like that, too (not just Stockhausen's music), and when there are similar shapes in different places, we may recognise their similarity, even when they are on quite different scales or occur in different transformations. The differently stretched formula segments in Licht can work in this way, though at some points the differences in scale may require an enormous listening effort to perceive.
I feel that it could be useful to have a discussion-forum on the music of Stockhausen. There are so many people from all over the world, young and old, learned and eager to get into contact with this musical world: musicologists, composers, musicians, music lovers; people who plan concerts - who write books or have to give lectures and so on. So there should be much stuff, many ideas that we can share. And when we have open questions, there may be people who studied just that and could give a hint or a stimulus.
A problem might be the English language, but i feel that is the only possibility that many people who are interested can participate. And we can exercise tolerance to mistakes!