Independent Component Analysis in FMRI is (usually) used to find a set of statistically independent spatial maps together with associated time courses. This is known as spatial ICA, and is used when there are more voxels of interest (i.e. those in the brain/cortex) than time points. If, on the other hand, there are more time points then it becomes possible to do temporal ICA.
These decompositions are well determined as long as the data consists of the mixture of a sufficient number of non-Gaussian signals. Therefore it is important to initially estimate the number of non-Gaussian signals and not attempt to estimate too many components, as then the decomposition is underconstrained.