A node is a point where two or more wires come together. If you place a point on a wire, that point becomes a two-wire node... a trivial construct. If you place a point where three (or more) wires are connected in common, then that becomes a three (or more) wire node. And so on, for as many wires as you care to add to that node. The whole point of KCL is that charge does not accumulate at nodes, so the currents in the wires connected to a node always algebraically sum to zero. Any node, any number or wires.
How can you determine if a wire is connected to a node? There will always be zero potential between that wire and the other wires connected to the same node. This all goes hand-in-hand with KVL, which states that the voltage drops around any circuit loop must algebraically add to zero. Together, they form a powerful analysis tool for series, parallel, and combinations of series and parallel circuits.
Throw in Ohm's Law and Thevenin Equivalent Circuits and you pretty much have DC circuit analysis in your tool bag.