Jns, there is an important aspect to this mitochondrial study that you have not addressed at all.
The study doesn't simply count the number of mitochondrial variants.
It also measures the genetic distance between them (this can be taken as a proxy for how long ago the two lineages diverged (whithin statistical fluctuations of course)).
And it turns out that all the lineages outside of Africa are more closely related to each other when compared with the lager variety found in Africa.
How do you explain that with your model?