Combining the Keldysh theory for a few cycle laser pulse with the theory of the inelastic tunneling (Kornev A S and Zon B A 2012 Phys. Rev. A 85 035402) agrees very well with the ab initio simulations performed by Rohringer and Santra (2009 Phys. Rev. A 79 053402) for the Ne atom. However, for the heavier Xe atom, agreement between analytical and numerical theories is not satisfactory. To eliminate this disagreement, we took into account the Stark shift of atomic level energies. As a result, the agreement between analytical and numerical theories significantly improved, but for Ne, it became worse. We discussed possible causes of these disagreements.