Hmm, I know how it feels to be new at this kind of thing. Defennder gave a good mathematical description of "curl B" in terms of matrices and determinants. Maybe it will help to describe in a hand-wavy fashion? It will be useful to you to draw a few field vectors on the X-Y plane to visualize the field.
The little carat over the x and y tells you that these are vectors. For example, the first part tells you that the field's y-component is ax, and the x-component is by. (like at (1,1), the B vector is <b,a> but at (-2,2) the vector is <2b,-2a>) If you do this, you will see the curly nature pop out.
"curl B" is a good descriptive term because the operation gives you a sense of how curly (or whirlpooly or whatever) the B field is. Static electric field lines tend to "diverge," or spread out, like the field lines from a point charge, so the mathematical operator \vec{\nabla}\cdot \vec{E}, called the "divergence" operator, sort of tells you how much the electric field spreads out. The dot product gives a scalar which is why the right side of Gauss's Law is just scalars (enclosed charge density & permittivity). Steady-current magnetic fields, on the other hand, don't diverge at all; rather, they loop in circles around the current. The curl operator \vec{\nabla}\times \vec{B} tells you how much the B field curls, and in what direction. That's an important aspect of the curl, it gives you a vector rather than a scalar. This should make sense; after all, current (or current density) has to have a direction, but static charge does not.
As to your question about the third dimension k, well, this is a consequence of the 3-D nature of magnetic fields. If you have a wire pointing up, you can call that the k (or z) direction, whereas the direction of the field lines is in the i and j (x and y) directions. Maybe do a google image search for "right hand rule" and "right hand curl rule current" for illustration of this 3-D nature.
So if you do the curl of your particular B field, you will get a vector function which tells you the current density and its direction. With a defined loop, you can figure out how much current is in the loop by integration. As defennder said, if you do the curl correctly you shouldn't get zero. Use your Calc textbook to help with the curl operation.