In view of the recent improved data on B_{s,d}->mu^+ mu^- andB_d->K^*mu^+mu^- we revisit two simple NP scenarios in which new FCNC currentsin b->s mu^+ mu^- transitions are mediated either entirely by a neutral heavygauge boson Z' with purely LH complex couplings Delta_L^{qb}(Z') (q=d,s) andcouplings to muons Delta_A^{mu mu}(Z') and Delta_V^{mu mu}(Z') or the SM Zboson with LH complex couplings Delta_L^{qb}(Z). We demonstrate how thesecouplings can be determined by future Delta F=2 and b->s mu^+ mu^- observables.The correlations between various observables can test these NP scenarios. Wepresent the results as functions of Delta M_q/Delta M_q^SM, S_{psi phi} andS_{psi K_S} which should be precisely determined in this decade. We calculatethe violation of the CMFV relation between BR(B_{s,d}-> mu^+ mu^-) and DeltaM_{s,d} in these scenarios. We find that the data on B_{s,d}->mu^+ mu^- fromCMS and LHCb can be reproduced in both scenarios but in the case of Z, DeltaM_s and S_{psi phi} have to be very close to their SM values. As far asB_d->K^*mu^+mu^- anomalies are concerned the Z' scenario can significantlysoften these anomalies while the Z boson fails because of the small vectorcoupling to muons. We also point out that recent proposals of explaining theseanomalies with the help of a real Wilson coefficient C^NP_9 implies uniquely anenhancement of Delta M_s with respect to its SM value, while a complex C^NP_9allows for both enhancement and suppression of Delta M_s and simultaneouslynovel CP-violating effects. We also discuss briefly scenarios in which the Z'boson has RH FCNC couplings. In this context we point out a number ofcorrelations between angular observables measured in B_d->K^*mu^+ mu^- thatarise in the absence of new CP-violating phases in scenarios with only LH or RHcouplings or scenarios in which LH and RH couplings are equal to each other ordiffer by sign.