We show that by extracting temporal and spectral features from EEG signal and, following, using neural network to classify those features, one can significantly improve the performance of Brain-Computer Interfaces (BCIs) in predicting which motor movement was imagined by a subject. Our movement prediction algorithm uses Sequential Backward Selection technique to jointly select the temporal and spectral features, and a radial basis function neural network for the classification. The method shows an average performance increase of 5.96% compared to state-of-the-art benchmark algorithms. Using two popular public datasets, our algorithm reaches 91.73% accuracy (compared to an average benchmark of 81.10%) on the first dataset, and 88.78% (average benchmark: 82.76%) on the second dataset. Given the high variability within- and across-subjects in EEG-based motion decoding, we suggest that using features from multiple modalities along with neural network feature selection and classification protocol is likely to increase BCI performance across various tasks.