If lean is ever going to become more mainstream people must start treating it less as a religion with its own gods and more as a collection of insights that have to be carefully tailored to the context you are working in.
Also "94% of the potential for improvement is in the system there is little point in working with organizational culture" is about the opposite of the point he was making. His point would better be stated that there is little point in working on "problems with individuals."
Working on "culture" can mean many things. And often it is just a big waste of time. But the reason for that is not due to systemic/common cause versus special causes.
What Dr. Deming was suggesting is you need to look for systems improvements (which could be cultural - a culture that operates with data based decision making, an understanding of variation...). I think the red bead experiment illustrates the point he was making - any focus on fixing the employees on that system is futile if you don't change the system.
Related: Blame the Road, Not the Driver - Common cause variation - Deming on Management