It's the same mechanism. At the end of the day it is driven by density differences; salinity and thermal. Both of which are temperature dependent. So if it is as you say orbital forcing is causing the planet to warm. Which according to the IPCC's graphics, isn't happening now so why would it in the past? According to me, it is the natural cycle of glaciation and deglaciation in the Arctic that is driving temperatures. As the northern hemisphere deglaciates the planet is returning to it's pre-glacial temperature. At some point density differences trigger a change in ocean currents and the cycle begins anew.
Ocean circulation is a redistributive mechanism, not an external energy source. AMOC shifts can strongly change where heat goes, especially regionally, but they do not explain the pacing of glacial cycles by themselves.
You're saying the Atlantic drives it, but that just moves the question back one level. What drives the long-term state changes in the circulation system itself?
The reason orbital forcing remains central is because the timing problem still exists.
The transition from dominant 41,000-year cycles to ~100,000-year cycles matches changes in ice sheet dynamics and orbital modulation.
Deglaciations consistently occur near peaks in northern summer insolation.
Ice volume, CO2, sea level, and temperature evolve in phase relationships predicted by orbital pacing plus feedbacks.
Ocean circulation alone does not explain that astronomical synchronization.
And yes, abrupt variability also occurred in some interglacials. Nobody disputes that. But instability within interglacials is not the same thing as explaining the existence and pacing of the glacial-interglacial cycle itself.
Also, you said...
"temperature changes alter density differences, which alter circulation."
Fine. But what initiated the large-scale temperature changes repeatedly over hundreds of thousands of years with orbital periodicities matching Milankovitch cycles?
If the answer is "internal ocean variability," then you need a mechanism explaining why that variability repeatedly tracks orbital geometry over geological timescales instead of behaving quasi-randomly.
And there's another issue...
The current warming pattern is not just North Atlantic redistribution. The oceans globally are accumulating heat. Measured ocean heat content has risen substantially. Redistribution can cool one region and warm another, but it cannot explain a long-term increase in total system heat without a forcing imbalance.
The IPCC graphic point cuts against your argument. Orbital forcing today is slightly toward long-term cooling, not warming. Yet temperatures are rising anyway. That's one reason greenhouse forcing stands out so strongly in attribution studies.