Take a look at the 'Matrix Functions' page in the GMS2 Manual. The basic idea is that the converter makes a vertex_buffer which you can think of as a 3D sprite. Then to draw that 3D sprite at a particular position you simply apply a transform (matrix transform to be precise) before drawing the buffer, that way you aren't having to rebuild the entire buffer every frame, which would be very inefficient.
Here is some sample code from a basic project I had when using this:
/*
(Draw Event)
var tempMatrix = matrix_build(x, y, z, 0, direction, 0, scale, scale, scale);
matrix_set(matrix_world, tempMatrix);
vertex_submit(my_buffer, pr_trianglelist, sprite_get_texture(tx_Earth, 0));
matrix_set(matrix_world, matrix_build_identity());
*/
'z' and 'scale' were both object defined variables. In the step event I simply had 'direction += 0.2' in order to make the globe I was rendering slowly spin. In order to move the object, you would just have to change the x, y, z, values in the 'var tempMatrix = matrix_build(x, y, z, 0, direction, 0, scale, scale, scale)' line.