I am familiar with the existence of the new blue star reeds. They do not influence what I said, because it wasn't about thick or thin reeds but thick or thin reed plates. It's during the time that the reed tip travels from one side of the reed plate to the other side that no air is flowing and the sound wave flatlines for a brief moment. That causes the distortion we know from old class B amplifiers (and to avoid that distortion in amplifiers class AB was created that keeps a bit of current flowing around the "null").
Class B gets a bad rap because of old unregulated class B amplification. In practice, the driver stage in solid-state amps has large negative feedback and travels across that gap faster than the speaker membrane can follow. My ancient Solton Turbojet from the seventies is class B (with a driver transitioning gap of about 2V) and you don't hear it in use. The idling noise does have a certain somewhat rustling character instead of only a constant hiss. And this is a transistor driver stage, with a single-stage differential amplifier, nowhere near the amplification factors of modern (or even old) opamps.
I also think your description is likely to create confusion about the main sound source in an accordion: the lion's share of the sound energy is
not created by the vibrating reed (which has lots smaller vibrating mass than a violin string, say) but by the compressed air stream chopped into bits by the vibrating reed. A significant contributor to a helicon bass reed's loudness is the trapezoid reed plate that prolongs the time that the reed blocks the air flow. Good reeds don't have all that much more vibration of their own than low-quality reeds, but they have minimal air gaps and thus cut off the air stream more completely.
That makes their tone more overtone-rich, meaning that reed chambers and cassotto and the grille and other acoustic factors get more material to work with. It also means that the direct tone quality is quite harsh, so there is actually more of a
need to mitigate it with cassotto and other measures. Cassotto instruments with machine reeds are more of a marketing gag than an asset, and the overtone structure does not have the strength where the cassotto imprint on it would make for a well-recognizable formant signature.
By the way: human voice also operates on the interrupted-air-stream principle with good closure (in lower vocal ranges, typically complete) and using mouth shapes on the overtone-rich material to convey vowels. Our ears are
trained to recognize formant structures (and tell speakers apart by their voices), so a cassotto instrument stands out. That is more of an asset for small ensembles than large orchestras, and it doesn't help if identically built instruments are scattered across the voice groups. The Russian "Timbre" ensemble utilizes this effect to the max.