alright look at my screenshot, and you can see the asymptotes are x = 3 and y = x - 4
now using addition of oordinates with these two doesnt yield that graph (i may be wrong). I assume that they have also graphed the hyperbola and used that as a substitute to x=3 when using addition of ordinates since how can you use x = 3 for this ? does this make better sense? hopefully haha 
Yes, I think I see your issue.
So, addition of ordinates is a tool used when you want to draw a function which is a sum of two other functions. Sometimes, this sum of functions has asymptotes - BUT you do not use the asymptotes to draw the function.
For example, take f(x)=x+2. We can consider this as a sum of two functions - f1(x)=x and f2(x)=2. So, we can draw y=x and y=2 on the same cartesian plane, and then use addition of ordinates to get an estimate on the shape of y=x+2.
Now, let's consider the harder example,
=x-4-\frac{1}{x-3})
. We can consider this as a sum of the functions
=x-4)
and
=\frac{-1}{x-3})
.
Now, to assist with us drawing, we would dot in the two functions
)
and
)
(as they've done). We've these two functions, we can now use addition of ordinates to get an estimate of the shape of the graph, remembering that these two functions aren't necessarily asymptotes.
After graphing it, you will notice that we do see asymptopic behaviour on the lines

and

, and so we would then dot those lines in to label them as asymptotes.
However, something with these graphs you can do is consider them in the general form
)
, where g(x) is some arbitrary function.
Now, imagine drawing the graph

. Next, what you do is sort of move that graph so that instead of it having asymptotic behaviour at y=0, it instead starts to asymptote towards g(x). (it's a little hard to explain, and really requires the help of a gif - but, I don't know how to make gifs, so we're at a loss there)
If you're still confused about why they drew the hyperbola, just say and I'll let someone else takeover. If you still don't understand, that's a statement about how I teach vs. how you learn, not how smart or dumb you are. (because really, we're all smart.

)