The majority of rain used by winter grain crops in the Mallee region of Victoria, Australia, falls during the cooler months of the year (April–October). However, rain falling during the summer fallow period (November–March) and stored as soil moisture contributes to grain yield. Strategies to better capture and store summer fallow rain include (i) retention of crop residues on the soil surface to improve water infiltration and evaporation; and (ii) chemical or mechanical control of summer fallow weeds to reduce transpiration. Despite the widespread adoption of no-till farming systems in the region, few published studies have considered the benefits of residue management during the summer fallow relative to weed control, and none quantify the impacts or identify the mechanisms by which summer fallow weeds influence subsequent crop yield. Over 3 years (2009–11), identical experiments on adjacent sand and clay soil types at Hopetoun in the southern Mallee were conducted to quantify the effect of residue management (standing, removed, or slashed) and summer fallow weed control (± chemical control) compared with cultivation on soil water and nitrogen (N) accumulation and subsequent crop yield. The presence of residue (2.4–5.8 t/ha) had no effect on soil water accumulation and a small negative effect on grain yield on the clay soil in 2011. Controlling summer weeds (Heliotropium europaeum and volunteer crop species) increased soil water accumulation (mean 45 mm) and mineral N (mean 45 kg/ha) before sowing on both soil types in 2 years of the experiment with significant amounts of summer fallow rain (2010 and 2011). Control of summer weeds increased grain yield of canola by 0.6 t/ha in 2010 and wheat by 1.4 t/ha in 2011. Using the data from these experiments to parameterise the APSIM model, simulation of selected treatments using historical climate data (1958–2011) showed that an extra 40 mm of stored soil water resulted in an average additional 0.4 t/ha yield, most of which was achieved in dry growing seasons. An additional 40 kg/ha N increased yield only in wetter growing seasons (mean 0.4 t/ha on both soil types). The combination of extra water and N that was found experimentally to result from control of summer fallow weeds increased subsequent crop yield in all season types (mean 0.7 t/ha on sand, 0.9 t/ha on clay). The co-limitation of yield by water and N in the Mallee environment means that yield increases due to summer weed control (and thus returns on investment) are very reliable.