
Each object has its own rotation, translation, and scaling  all are optional, you can have all three, one or two, or none.
You don't say you are using DirectX or OpenGL  but the basics are the same for both APIs. The "world transformation" is nothing more than these operations, set for each object. Let's say you want to rotate and position each object. Each object has three rotation angles, we'll call them XROT, YROT, and ZROT, expressed in radians. And each object has a three position coordinates we'll call XPOS, YPOS, and ZPOS.
What you need to do is get these values into matrices. Look at your spinning triangle demo. Somewhere in there you will see the rotation values being placed into a matrix via some function call. Position can also be placed in a matrix. You multiply the rotation matrix by the position matrix, and store the result in the "world transformation" matrix. All the world transformation matrix is, is a "concatenated" matrix  the rotation, position, and possibly the scaling matrix for the object multiplied together.
So  each object has IT'S OWN world transformation matrix. Just set this matrix to the world transform before you draw the object.
I suggest you get a good basic book on 3D programming and study it and work through the examples  that's the way to learn this stuff. A book will help you build your knowledge stepbystep much better than online tutorials will. Once you have some basic knowledge, then use the online tutorials.
Many people putdown the PrimaTech books, but for beginners they can be a good introduction. If you are using DirectX look at "The Zen of Direct3D Programming" and "Special Effects Programming with Direct3D".
You MUST understand these concepts to get anywhere with 3D programming:
= Rotation Angles = 3D coordinate space = matrix use, and the underlying theory of how matrices work = vector math
Good luck!
tAE
