I think I understand what you’re saying, but I don’t feel like it’s a flaw, more a design decision that can be played with.
Just to make sure you understand how the knobs function, when you have a signal plugged into the fall cv input, the knob becomes an attenuator for that signal - lets say we plugged in a 0v to 8v LFO - when the knob is fully counter-clockwise, you get a full decay, with no influence from the LFO. Then, as we turn the knob clockwise, we get more and more LFO modulation, which will in practice push the decay length shorter as higher voltages are reaching the CV. Once we get fully clockwise, the LFO has full control over the parameter, so there will be wide fluctuation. The only time you’ll get “full length” decay with this level of cv control is if the input is at/near 0v.
The rise cv input is a flipped version (e.g. fully counter clockwise you get no attack or cv).
This is why I recommend starting by patching gates to the CV inputs, at least to start with - gates in Lorre Mill are made up of 0v and 8v alternating signals. What this means is that, no matter where the fall knob is, you’ll get some full length decays from the low gates and then the high gate decay length will be determined by wherever you put the knob (so, if you wanna alternate between the longest and shortest decay time, you’d just go full clockwise, if you want the high gate to be halfway you put the knob halfway etc). It’s a lot easier to get long decays that way!
Also, I think I mention this in my Mosstone video, but a fun trick is to send the same gate to both the rise and fall CV inputs with a banana cable and play with the knobs - the result will be that, when the gate is high, the attack will be long and the decay will be short, and then when the gate is low, the attack will be short and the decay will be long - flip flopping envelopes!
Also, patching the envelope output back into the rise or fall CVs does have a subtle but noticeable effect depending on where the knob is, especially if you have the envelope modulation the fold/FM as well. The changes will be as follows:
Env Out -> Rise CV = Adjustable Logarithmic Attack
Env Out -> Inverter In, Inverter Out -> Rise CV = Adjustable Exponential Attack
Env Out -> Fall CV = Adjustable Exponential Attack
Env Out -> Inverter In, Inverter Out -> Fall CV = Adjustable Logarithmic Attack
What I mean by “Adjustable” is that the curve will become increasingly present the further clockwise you have the corresponding knob - fully counter-clockwise, the envelope’s curve will be linear, and then it will start to warp as you turn the knob clockwise.
Hopefully this helps and isn’t too confusing, lemme know if you have any other questions!