An analytical model based on global load sharing (GLS) theory is developed to guide design of fiber-reinforced hybrid composites with superior mechanical properties to single-fiber-type composites. The hybrid is assumed to comprise two types of fibers-low- and high-elongation-with sufficiently different failure strains. Hybridization is found to be most beneficial when small-to-moderate volume fractions of the low-elongation, high-strength/stiffness fibers are added to the high-elongation composite. In this regime, all key properties of the hybrid composite can be maintained or improved, relative to the pure high-elongation composite. Potential gains in stiffness are large (approximate to 50%), and gains in pullout stress are moderate (approximate to 10-30%), while failure strain is maintained by design. Furthermore, using discontinuous low-elongation fibers improves hybrid composite performance because such fibers fragment more gracefully over a wide range of strain. Because composite processing might be easier when both fiber types are discontinuous, the performance of hybrids using discontinuous high-elongation fibers are also investigated, and good performance can be largely maintained if the high-elongation fibers are sufficiently long. The analytical model is supported by exact GLS results, and is thus a useful design tool for developing higher-performance composites by hybridization. (C) 2015 Elsevier Ltd. All rights reserved.