Finally success!

I created a data-driven approach in extracting records from an Excel (xlsx) file into a cursor/dbf file.
1. I create an XML file (excel2dbf) of the following format:

<?xml version="1.0" encoding="utf-8" ?> 
<FIELDS>
        <Field>
                <Column>1</Column>
                <FldName>Fld1Name</FldName>
                <FldType>C</FldType>
                <FldLen>30</FldLen>
		<FldDec>00</FldDec>
        </Field>
        <Field>
                <Column>2</Column>
                <FldName>Fld2Name</FldName>
                <FldType>C</FldType>
                <FldLen>15</FldLen>
		<FldDec>00</FldDec>
        </Field>
        <Field>
                <Column>3</Column>
                <FldName>Fld3Name</FldName>
                <FldType>N</FldType>
                <FldLen>12</FldLen>
		<FldDec>02</FldDec>
        </Field>
        <Field>
                <Column>4</Column>
                <FldName>Fld4Name</FldName>
                <FldType>D</FldType>
                <FldLen>8</FldLen>
		<FldDec>00</FldDec>
        </Field>
       ... other fields
</FIELDS>

The value in the tag Column corresponds to the Column of the spreadsheet. So that, when extracting records from xlsx I can just specify the columns I need to have.
The PRG opens the XML, creates a cursor/table and then - using Greg Green's XLSXWorkBookVFPX - extracts records and populates the target cursor/DBF.
Therefore I will be able to use it on future projects.
Thank you, Greg, and others for your help!