The ability to predict crop yield during the growing season is important for crop income, insurance projections and for evaluating food security. Yet, modeling crop yield is challenging because of the complexity of the relationships between crop growth and the interrelated predictor variables. Artificial neural networks (ANNs) are useful for such complex systems as they can capture non-linear relationships of data without explicitly knowing the underlying processes. In this study, an ANN-based method (Advangeo® Prediction Software) was used to evaluate: 1) the relative importance of predictor variables for corn and soybean yield prediction, and 2) the potential of ANNs for predicting corn and soybean yield. Several satellite derived vegetation indices (e.g. normalized difference vegetation index - NDVI, red edge NDVI, simple ratio - SR, and the land surface water index - LSWI) and slope data were used as crop yield predictor variables, hypothesizing that different vegetation indices reflect different crop and site conditions. The study identified the SR index and the slope as the most important predictor variables for both crop types during both years. The number and dates of the images however were different for the two crop types (earlier dates for corn) and for the wetter (2011) and drier (2012) years. The relative mean absolute errors (RMAEs) were overall smaller for corn compared to soybean and 100% of the corn study sites had errors below 20% in both years. The errors were more variable for soybean. The results are promising and can provide yield estimates at the farm level, unlike current county level approaches.