Interesting example! Especially because the steering element of these motors is so counter intuitive - you push left to go right and vice versa :-)
I agree with you about the direct connection between the user's movements and the speed, but the direction of the twist required to go forward/backward is fairly arbitrary (it seems), so it must be learned rather than 'intuited'.
Something else to consider - many of the newer BMW designs have moved to the 'standard' approach. Why make the move, given the 'brand' value already mentioned and the history behind the design?
3 reasons come instantly to mind - consistency, human physiology and safety. :-)
I agree with most of what you're saying here, particularly the historical roots of the design and the post does acknowledge the importance of the interface the BMW Motorrad brand.
But I think you make a false assumption about the potential order of operations - I'm not worried about signaling a lane change when I'm in the middle of an emergency, I'm worried about the emergency occurring when I'm in the middle of signaling a lane change. :-)
Under normal circumstances, one should not brake and turn. But neither should cage drivers turn left in front of you. Or merge into your lane. Or run yellow lights. This stuff happens on the road, and this why so many riders get hurt.
To think that the normal sequence of events is the only one worth designing for is naive.
The post was about how to consider the human context when designing, and used the bike as an example. Any design decision that could delay a rider's reaction time to the primary brake is a bad idea, regardless of how familiar you are with your bike or how adaptable you may be.