Future accelerator magnets will require a large amount of high-performance Nb3Sn superconductors. However, the performance of present Nb3Sn candidates – mainly rod-restack-process (RRP) and powder-in-tube (PIT) strands – still falls short of the needs. Moreover, the critical current density (Jc) of Nb3Sn superconductors has been plateaued over the last decade. In this talk I will present several novel schemes to optimize the design of Nb3Sn strands, which make further significant improvement of their performance possible. These new conductors, which not only have higher Jc, but also promise to significantly reduce the cost, are promising to fabricate more powerful magnets.