Timo,
First I can understand Brien a bit, since indeed the whole history feels a bit chaotic to me as well.
If I see these sketches and pictures, I do have the feeling some basics can't be based on advice you got here or read in your books.
I find it difficult to solve like this.
It's not easy to say what the impact of all this different things can be.
As per your sketches you nowhere used mineral wool in the cavities or your walls and ceiling. Is that correct?
Basically you NEVER build double leaf walls like that (when related to sound isolation) without mineral wool or whatever absorption material in them.
Any double leaf wall system will show a resonance (which you described yourself already, since you heard it shift after you glued whatever to it.)
Having no absorption in that cavity enhances the resonances in that wall (that absorption acts as a demper).
Therefore you probably will faster hear differences by adding something as damper (gluing the tiles against these walls). The worse something is the faster you will hear improvements.
You used these tiles wrongly by gluing them and additionally painting them. Caulking them you either did for the visible effect, or you misunderstood entirely why you mounted them there.
If only open cell absorption was involved, I should think you neutralized their functionality in a significant manner, which also means that they can't harm your project, which is the good side of it.
But depending of how you glued them it's possible they can still resonate as a membrane.
These tiles are also not meant to create a sandwich panel between drywall (I wonder where you got that idea) but i think it doesn't matter that much.
In order to judge these Armstrong measurements (that second one glued against the hard backing) I should need to feel and see these boards and know exactly how they were glued and mounted.
For now I have the feel that other factors are involved than just gluing these boards against a rigid heavy backing.
But this doesn't matter that much anymore, since you have what you have, and mounted and painted them as you did.
If you use corner absorption it's certainly useful to use better absorbers than the LENRDs since they are rather small and therefore more limited in the low frequency absorption. This is not mainly related with the fact that they are foam but that they are that small.
I saw that you also experimented with fibreglass or rockwool absorbers? Did they gave the same effect? Normally, if they're OK they should go lower in frequency than the LENRDs.
You don't need to loose the LENRDs since you can use them as cavity filling behind these DIY absorbers stradling the corner.
These polys you made will also show a resonant behavior. Have you tried filling the back with wool?
I find your project a bit overwhelming, and find it hard to analyze all possible influences like this.
Have you calculated room modes to see if you find some relationship with your measurements?
At a certain moment, as Brien said, it can be more efficient to involve a local expert/acoustician, not via the phone for a detail here, then there, but directly on the spot, and let him check this whole thing.
To be honest I get the feel that this is a typical story where trying to do everything yourself wasn't the most cost, time and energy effective approach.
DIY is cost efficient if from the start you follow known, described and proven methods, and limit experiments (which can be fun, useful and educational) to very specific things in a controlled manner.
However if you want to continue then just forget a lot of things and treat your room by adding good bass traps in the corners (enough of them) based on your DIY absorbers and use your LENRDs as cavity filling, hang such absorbers on the reflection points and so on.
To get an idea look at the related layouts (which are all about the same) of different suppliers at their respective sites:
- Ready Acoustics
- GIK Acoustics
- RPG
- RealTraps
- Others?
And start from there.
Further:
Your earlier question of drilling or making holes in these walls to let them act as Helmholtz resonators.
Indeed they should act as resonators causing narrow band absorption (even more so since there is no wool in these walls).
Simultaniouly they will
significantly diminish the transmission loss of these wall at these very same frequencies.
With the uncertainties involved (also in tuning them) and the fact that you permanently damage your walls it's almost a disaster in the making.
For your own sake: don't.
I can't judge from here if it's sensible trying to fill these wall cavities with blow-in wool or whatever.
Success
, Let's continue the story....
I am an obsessed cheater, user names switcher, desperate illiterate, in need for urgent professional help ...?? 
