As pointed out by others, the answer to this depends on how IT is viewed within the company. In most companies IT is seen as a necessary evil, and consequently understaffed.
In that situation you need to sit down with your boss and present him with a list of all the stuff that needs/wants doing, complete with priorities from your point of view. This must include an allowance to deal with helpdesk issues (which you should arrive at by measuring how much time you are actually spending on it).
The list should also have estimates on how long all the other things need to implement or complete.
You know how much time you (and your team mates) have, so from there it is a simple exercise. Cut the allowance for helpdesk stuff out from the available time, which gives you remaining time. The boss provides the priorities and from there you can see how long it takes to implement the rest.
By the sounds of it, the picture will be extremely ugly. Which is exactly what you want, because that is then the time to say to your boss: "Well, if you want me to deliver more, then I need more resources. Of course, if IT cannot have more resources, then we will follow your priorities, which means that the entire bottom of the list never gets done.".
And now to the question of balancing helpdesk vs. infrastructure projects: In any engineering discipline (and IT is no different to that) there is well proven motto: Fix your problems twice.
First you fix the problem at hand, which is generally a symptom of something else being wrong. This makes sure the user is happy and goes away.
Then you fix the underlying root cause of the problem. That makes sure the symptoms don't come back.
If you do this with perseverance and consequence, you will end up in a situation where the helpdesk calls get less and you have more time for other stuff (until your bosses decide that you are overstaffed and cut your team size down).