I would doubt this is much more than (as has been noted) putting a layman filter over some stats.
My 'Big Book of Caveman' put mitochondial Eve at 200,000 BCE. It also states (on a particularly colourfull pop up page) that at about 150,000 there was a bottleneck in the gene pool as the popn fell to a few 10s of thousands.
Pushing this forwards to about 5-7k simply by plugging an arbitary number into a computation seems foolish.
This coupled with the logical arguement about migration featured above makes my think this artical is not saying what it is puported to.