Of course, all models are wrong in as much as they simplify reality, but my opinion is that this model is not even useful _for understand the dynamics of Americ

Of course, all models are wrong in as much as they simplify reality, but my opinion is that this model is not even useful _for understand the dynamics of American segregation_. It's interesting and probably has other legitimate applications in understanding real-life social phenomena, high school lunchroom seating segregated into jocks, geeks, stoners, and goths perhaps. But it's very misleading as a metaphor to describe American housing segregation.