In this article, the problem of 3-D target localization from bistatic range measurements in multiple-input-multiple-output (MIMO) radars with distributed antennas is considered. A closed-form two-stage weighted least squares solution is developed, which is able to uniquely determine the target position with a smaller number of sensors (transmitters or receivers) compared to the other existing closed-form methods. The proposed method employs several weighted least-squares minimizations and does not require any initial solution guesses to give the target position estimate. A detailed theoretical performance analysis associated with the method as well as the Cramer-Rao lower bound (CRLB) are provided. The proposed estimator is shown theoretically to achieve the CRLB accuracy under small Gaussian noise. Computer simulations are included to corroborate the theoretical development and to compare the estimator performance with several previous approaches. The proposed algorithm in the considered example is shown to outperform the existing closed-form estimators. In the 3-D and in the case of a minimum number of sensors (i.e., two transmitters and two receivers) in a MIMO radar, the proposed method is also shown to be able to localize a target with high accuracy while the previous closed-form methods cannot uniquely determine the target position.