This is mentioned in the article, I must admit I previously thought it was as you say, but:

"available evidence in humans suggests that hyperinsulinemia precedes weight gain (rather than the other way around)"