Constituent Quark Model versus Nonperturbative QCD

We discuss a few, apparently different (but actually, tightly related) problems: 1. The relation between QCD and valence quark model, 2. The asymptotic behavior of the nonperturbative pion wave function $\psi(\k, x)$ at $x\rightarrow 0,~1, \k \rightarrow \infty$ and 3. The dimensional counting rules in the intermediate region of energy. The analysis is based on such general methods as dispersion relations, duality and PCAC. We calculate the asymptotic behavior of the wave function ($wf$) at the end-point region ($x\rightarrow 1$ and $\k\rightarrow\infty$) by analysing the corresponding large $n-$th moments in transverse $\la\vec{k}_{\perp}^{2n}\ra \sim n!$ and longitudinal $\la (2x-1)^n\ra \sim 1/n^2$ directions. This information fixes the asymptotic behavior of $wf$ at large $\k$ ( which is turned out to be Gaussian commonly used in the phenomenological analyses). We discuss some applications of the obtained results. In particular, we calculate the nonleading ``soft" contribution to the pion form factor at intermediate momentum transfer. We argue, that due to the specific properties of $\psi(\k, x)$, the corresponding contribution can temporarily {\bf simulate} the leading twist behavior in the extent region of $Q^2$. The same effect also takes place for the nucleon formfactor. Such a mechanism, if it is correct, would be an explanation of the phenomenological success of the dimensional counting rules at available, very modest energies for many different processes. We discuss some inclusive amplitudes also.