The 'air molecules impinging on the airfoil' theory of lift is not wrong, but does not account for a large fraction of the lift generated either. In a dense enough medium such as air at normal flight altitudes, you have to take into account how those air molecules interact with other air molecules and not only the airfoil. Sum up all those interactions, and you get fluid dynamics rather than a molecule hitting a surface in vacuum.