I'm not sure why nobody has mentioned this yet, but the situation becomes clear when you assign R and S configurations to each stereocenter. A stereocenter is defined as the position of an atom with four unique substituents. Also, lone pairs of electrons are substituents, so don't skip over an atom on which you only see three atomic substituents. This usually happens when nitrogen is locked in a cyclic structure.
What you should do in all optical-activity-type questions is first find and label all the stereocenters as R or S. If such a designation can be given to an atom (for example, carbon), that atom is chiral. For this reason, stereocenters are more commonly called chiral centers. Read the rest of the post after you have learned how to make these designations.
First, assume both molecules are different and nonsuperimposable. One is S,R and the other is R,S. As you should already know, when there are chiral centers in a molecule, the molecule should rotate plane-polarized light.
There is a special case when symmetry ruins this property: when the two molecule are actually the same. Important cases like this deserve a name, a meso compound. Meso compounds do not optically rotate light. For your problem, it's really obvious that the molecules are actually the same, because they're drawn as Fischer projections. In a Fischer projection, the side groups are out of the plane, and the "backbone" is "into" the plane. Rotating a Fischer projection does not change the fact that the side groups are still out of the plane. You should really look up Fischer projections to see why. You can see some good examples if you search Google Images.
It's more difficult to determine whether a molecule is a meso compound if it's not drawn as a Fischer projection. In the case that a molecule is drawn as a dash-wedge structure, one should attempt to recognize if it is a meso compound by exploiting symmetry instead of trying to superimpose images. However, exercise caution. As you have said yourself, planes of symmetry don't always denote meso compounds. You have to first find out if there are chiral centers in the molecule at all. I'm guessing what your book says is another caution: planes of symmetry can be difficult to find.