Mean pressure gradient affects the turbulence mainly through the modulation of the mean rate of strain. Modification of the turbulence structure feeds, in turn, back into rite mean flow. Particularly affected is the near wall region (including the viscous sublayer) where the pressure gradient invalidates the conventional boundary-layer "equilibrium" assumptions and inner-wall scaling. Accurate predictions of such flows require application of advanced turbulence closures, preferably at the differential second-moment level,with integration Lip to the wall. This paper aims at demonstrating the potential usefulness of such a model to engineers by revisiting some of the recent experimental and DNS results and by presenting a series of computations relevant to low-speed external aerodynamics. Several attached and separated flows, subjected to strong adverse and favorable pressure gradient, as well as to periodic alternation of the pressure gradient sign, all computed with a low-Re-number second-moment closure, display good agreement with experimental and DNS data. It is argued that models of this kind (in full or a truncated form) may serve both Sol steady or transient Reynolds-Averaged Navier-Stokes (RANS, TRANS) computations of a variety of industrial and aeronautical flows, particularly if transition phenomena, wall friction, and hear transfer are in focus.