Robot kinematics modeling has been one of the main research issues in robotics research. For realtime control of robotic manipulators with high degree of freedom, a computationally efficient solution to the inverse kinematics modeling is required. In this paper, an SOM-Like inverse kinematics modeling methodis proposed. The principal idea behind the proposed modeling method is the use of a first-order Taylor series expansion to build the inverse kinematics model from a set of training data. The work space of a robot arm is discretized into a cubic lattice consisting of Nx×Ny×Nz sampling points. Each sampling point corresponds to a reciprocal zone and is assigned to one neural node, storing four different data items(e.g., coordinates position vector, template position vector,the joint angle vector, and the Jacobian matrix) about the first-order Taylor series expansionof the inverse kinematics function at that sampling point. The proposed inverse kinematics modeling method was tested on a 3-D printed robot arm with 5 degrees of freedom (DOF). The performance of the proposed method was tested on two simulated examples. The average approximation error could be decreased to 0.283 mm in the workspace, 200.0 mm×200.0 mm×72.0 mm and 0.25 mm in the workspace, 200.0 mm×200.0 mm.