We use our numerical code , DRAGON , to study the implications of recent data on our knowledge of the propagation properties of cosmic ray nuclei in the Galaxy . We show that B/C ( as well as N/O and C/O ) data , including those recently taken by CREAM , and { \bar { p } } / p data , especially including recent PAMELA results , can consistently be fitted within a unique diffusion-reacceleration model . The requirement that light nuclei and \bar { p } data are consistently reproduced within experimental uncertainties places significant limits on the main propagation parameters . In particular , we find the allowed range of the diffusion coefficient spectral index to be 0.3 < \delta < 0.6 at 95 % confidence level and that Kraichnan type diffusion is favored with respect to Kolmogorov . While some amount of reacceleration is required , only a limited range of the Alfvèn velocity value ( 10 \raise 1.29 pt \hbox { $ < $ \kern - 7.5 pt \raise - 4.73 pt \hbox { $ \sim$ } } v _ { A } % \raise 1.29 pt \hbox { $ < $ \kern - 7.5 pt \raise - 4.73 pt \hbox { $ \sim$ } } 20 ~ { } { km } ~ { } % { s } ^ { -1 } ) is allowed by a combined analysis of nuclear and antiproton data , which we perform here for the first time . If antiproton data are not used to constrain the propagation parameters , a larger set of models is allowed . In this case , we determine which combinations of the relevant parameters maximize and minimize the antiproton flux under the condition of still fitting light nuclei data at 95 % C.L . These models may then be used to constrain a possible extra antiproton component arising from astrophysical or exotic sources ( e.g . dark matter annihilation or decay ) .