[en] Recent data on Galactic cosmic-ray (CR) leptons and hadrons gave rise to two exciting problems: on the lepton side, the origin of the rise of the CR positron fraction e ${}^{+}$/(e ${}^{-<!--\; ???\; -->}$ + e ${}^{+}$) at ∼10–300 GeV of energy; on the hadron side, the nature of the spectral hardening observed in CR protons and nuclei at ∼TeV energies. The lepton anomaly indicates the existence of a nearby $e^{\pm <!--\; ??\; -->}$ source. It has been proposed that high-energy positrons can be produced inside nearby supernova remnants (SNRs) via interactions of CR hadrons with the ambient medium. A distinctive prediction of this mechanism is a high-energy rise of the boron-to-carbon ratio, which has not been observed. It also requires old SNRs at work (with ineffective magnetic field amplification and slow shock speed) that cannot account for the CR hadronic spectra observed up to the knee energies (∼5 PeV). We propose a new picture where, in addition to such a nearby CR accelerator, the high-energy spectrum of CR hadrons is provided by the large-scale population of SNRs, younger on average, which can efficiently accelerate CRs up to the knee. Under this scenario, the spectral hardening of CR hadrons can be naturally interpreted as the transition between the two components. As we will show, our two-component model breaks the connection between the positron fraction and the boron-to-carbon ratio, which is now predicted to decrease with energy in accordance with the data. Forthcoming data from AMS will be crucial for testing this model.