C#

public class Storage
{
    int value;
}

public class Test
{
    public Test(Storage val)
    {
        m_val = val;
        m_val.value = 10;
    }
    private Storage m_val;
}

public class Test
{
    public Test(){}

    public int MyNumber
    {
        get{return m_i;}
        set{m_i = value;}
    }

    public int m_i;
}

public class SomeOtherClass
{
    public SomeOtherClass(){}


    public void ChangeValue(ref int i)
    {
        i = 5;
    }
}

    // Example
Test t = new Test();
SomeOtherClass soc = new SomeOtherClass();
t.MyNumber = 8;
soc.ChangeValue(ref t.m_i);
Console.Write(t.MyNumber);

1. Internet Explorer uses the WebBrowser object, so there really is no difference. The InternetExplorer automation server is just an out-of-process container for the WebBrowser control (which hosts and provides services to MSHTML). So no, it won't fetch pages any faster.
2. Settings for the WebBrowser control (and many affect MSHTML in general) are global. Only a few things like printer settings can you override on a per-instance basis. You can use the hosting interfaces, however, to control a few aspects of the WebBrowser control, such as the context menus, URL resolution when a link is clicked, and several other things. See Using MSHTML Advanced Hosting Interfaces[^] for more information.