You've probably seen the railway network diagrams that big cities have at their stations and stuff. These are stylised representations of the actual rail network, and serve a specific purpose: to show commuters which line takes you to which destination very easily. They are not drawn to scale, and bear only a passing resemblance to geography, though they are topologically correct (i.e. a loop is still a loop, a branch is still a branch). Detail that does not help the commuter (or, for that matter, a train scheduler) is removed.
A circuit diagram does the same thing. To allow a technician to understand the components and connections in a circuit, they are drawn on a paper in a way that shows these and their relationships to each other clearly. To represent them as they are in an assembled unit may be possible, but may be extremely difficult to follow (specially as units are three dimensional and paper diagrams are not).
If you build a component (e.g. an amplifier) from a kit, you will get two pieces of information: the circuit diagram that shows how the thing works, and a wiring diagram to show you how to put it physically together. These days, greater use is made of large-scale integrated chips, so many functions are delivered within a single chip, and all that attaches to a circuit board is the hardware (e.g volume knob). The circuit board itself is an exercise in topology, because it is in effect a two-dimensional object, and thus necessity may thwart a desire to have the board mimic a workflow.
In any case, where components are flown from the board, the placement of these can be quite arbitrary (i.e. you could position an indicator led anywhere you like on a panel).
(Seeing as deavers23 class should now be over, I figure I can be more blatant than Glen's subtle prodding)