A word about mappings in general:
An element $(a,b)$ of a mapping $f:A \to B$ is a pair. We say: $f$ maps $a$ to $b$. the set $A$ is called the DOMAIN (or souce set) of $f$, and the set $B$ is called the CO-DOMAIN (or target set).
You can think of $f$ as something that "grabs" elements of $A$, and "throws" them into $B$. Two elements of $A$ may hit the same target, but each domain element can only be thrown "once".
The "$b$" of a pair $(a,b) \in f$, is called the IMAGE of $a$ under $f$, and we often write $b = f(a)$ (Apparently your text writes $b = (a)f$. This is non-standard, but some authors do it, so that composition of functions occurs in the same order we read: left-to-right). It's important to realize that just listing the RELATIONSHIP of $b$ to $a$ is "not enough", for example it is bad to write:
"the function $x^2$"
and better to say, the function $f:A \to B$ such that $f(a) = a^2$ (or in your "style," $(a)f = a^2$) for every $a \in A$ (of course, this pre-supposes that $a^2$ "makes sense" for $B$, that is, we can multiply elements of $A$ together, and such multiples are, in fact, in $B$).
Another way mappings are often indicated is like so:
$f: a \mapsto f(a)$ (or, again, $a \mapsto (a)f$), but this again, is "not enough", we have to say WHAT SETS (what kinds of "things") $a$ and $f(a)$ (or...$(a)f$...) live in.
With your fourth example, we have the domain: $J \times (J-\{0\})$, whose elements are pairs of integers, with the second element of the pair non-zero. For example, one such element is $(-4,3)$.
Under the mapping $\tau: J \times (J - \{0\}) \to \Bbb Q$ we have:
$(k,m)\tau = \dfrac{k}{m}$, for example, with the pair above, we obtain $(-4,3)\tau = -\dfrac{4}{3}$.
Note that this mapping takes the pairs $(2,2)$ and $(3,3)$ (which are clearly "different" pairs) to the same rational number, $1$.