I'm assuming that these aren't priority queues - just items in whatever order they've been added into the queues. So here's how to go about it.

# Start with an empty master queue.
my @master; 

# Build a sub queue. Note that this is a array *reference*.
my $subqueue = [ 1, 2, 3, 4];

# add it to the queue. push() puts it at the tail of the queue.
push @master, $subqueue;

# Now let's create a couple more sub queues and add them.
push @master, [5, 6, 7];

# Some high-priority-items we'll stick at the front of the master:
unshift @master, [-1, 0];

# Now let's run the queue. We look at the master. If there are no item
+s remaining,
# the queue is empty. Otherwise, process the leading sub queue.
while (@master) {
  # If the leading sub queue is empty, discard it, and start over with
+ the rest of
  # the (now possibly empty) master queue.
  while  ( @{ $master[0] } ) {
      my $item = shift @{ $master[0] };
      print $item, " ";
  }
  # We arrive here when the sub queue is empty. Discard the empty sub 
+queue.
  shift @master;
  # Move to the next line so we can see we switched sub queues.
  print "\n"
}
print "\n";
[download]

The output will be

-1 0 
1 2 3 4 
5 6 7
[download]

Obviously this is a toy program, but it serves to demonstrate the basic array operations and tests that are needed to manage a queue of queues. More complex operations (like priority queuing, etc.) are left as an exercise. (I'd probably switch the plain scalar values for anonymous hashes, one item of which is a 'priority' field that you can sort the array of hashes on, and the other a 'value' which will be either the scalar value you want in the node (on the sub queues) or a reference to a subqueue (for the master).)

And of course this isn't a class; that'd be a nice thing to do as well.